- Email: [email protected]
Elevated plasma cholecystokinin and appetitive ratings after consumption of a liquid meal in humans
PRELIMINARY REPORT
Elevated Plasma Cholecystokinin and Appetitive Ratings After Consumption of a Liquid Meal in Humans Laurence J. Nolan, PhD, Janet L. Guss, MS, Rodger A. Liddle, MD, F. Xavier Pi-Sunyer, MD, and Harry R. Kissileff, PhD From the New York Obesity Research Center, St. Luke’s–Roosevelt Hospital Center and Columbia University College of Physicians & Surgeons, New York, New York, USA; and the Department of Medicine, Duke University Medical Center, Durham, North Carolina, USA OBJECTIVE: This study had two objectives. The first was to evaluate the possibility that, in a previous study, a soup preload augmented the reduction of food intake in a test meal induced by an exogenous infusion of cholecystokinin (CCK) because the soup also endogenously released CCK. The second was to compare CCK release by soup between men and women to determine whether the increased satiating effectiveness of soup in women as opposed to men could have been partly attributable to differences in CCK release. METHODS: By using a bioassay that measures all of its known isoforms, we determined plasma CCK levels at baseline and at several times postprandially in eight healthy, non-obese men and women (four of each sex). Each subject ingested 800 g of tomato soup, which was followed 30 min later by 300 g of a yogurt shake. Appetitive ratings were also collected and related to CCK levels. RESULTS: Ingestion of tomato soup significantly increased plasma CCK levels by 3.81 pmol/L (⫾ 1.21 standard error, P ⫽ 0.016) over baseline within 30 min in all subjects combined. When CCK concentrations at 5 min after soup and 5 min after yogurt were averaged, the women’s mean averaged concentration was 5.58 pmol/L (⫾ 1.994, t ⫽ 2.80, P ⫽ 0.0073) higher than the men’s. The elevated levels persisted but did not rise further upon consumption of the yogurt shake. Hunger ratings declined and fullness ratings increased after eating, although patterns of ratings did not match exactly patterns of CCK release. CONCLUSIONS: A large quantity of tomato soup stimulates significant CCK release; therefore, some of the satiating effects of soup preloads could have been mediated by an elevation in endogenous CCK. Nutrition 2003;19:553–557. ©Elsevier Inc. 2003 KEY WORDS: cholecystokinin, human ingestive behavior, satiety, eating behavior
INTRODUCTION The influence of cholecystokinin (CCK) in the control of food intake has been investigated for almost 30 y, but its role in normal eating episodes in humans has yet to be fully explored. The evidence for activity of CCK in the control of food intake includes the following facts. The exogenous administration of CCK reduces food intake in all species tested, including humans (for review, see Smith and Gibbs1), and CCK antagonists increase food intake in animals (for review, see Smith and Gibbs1). After the demonstra-
This study was supported in part by the following grants from the National Institutes of Health: DK36507, DK53089, and T32 DK07559 to St. Luke’s/ Roosevelt Institute for Health Sciences (H. R. Kissileff, principal investigator); DK38626 to Duke University Medical Center (R. A. Liddle, principal investigator.); and DK26687 to The New York Obesity Research Center (F. X. Pi-Sunyer, Director). Laurence J. Nolan’s current address: Department of Psychology, Wagner College, Staten Island, NY, 10301, USA Correspondence to: Harry R. Kissileff, MD, St. Luke’s-Roosevelt Hospital Center, 1111 Amsterdam Avenue, New York, NY 10025, USA. E-mail: [email protected] Nutrition 19:553–557, 2003 ©Elsevier Inc., 2003. Printed in the United States. All rights reserved.
tion by Moran and McHugh,2 that a saline preload is necessary for a low dose of CCK to reduce food intake in monkeys, our laboratory showed that reduction in food intake induced by exogenous administration of CCK in humans is enhanced significantly by combining it with a preload of soup.3 Post-meal release of CCK has been demonstrated in humans4 – 6 and correlated with verbal ratings of satiety.7 Proteins and fats are the best macronutrient secretagogues for CCK in humans.8 –10 Antagonists specific for the CCK-A receptor have been shown to reduce the satiating potency of intestinal fat infusions (loxiglumide)11 and increase hunger ratings (devazepide).12 Further, loxiglumide blocked the effect of CCK to suppress food intake when combined with preload of bananas and water.13 Conversely, French et al.14 reported that loxiglumide has no effect on hunger ratings. In three studies after administration of loxiglumide, food intake (in units of weight) in humans11,14,15 did not increase significantly, although energy intake increased significantly by 10% in one study.15 Thus, these studies showed a role for endogenous CCK in sensations of satiety in humans, although its magnitude and the essential conditions for its effectiveness remain elusive. The results of a previous study3 suggested that a nutrient-dilute liquid food such as tomato soup augments satiety induced by an intravenous CCK infusion, mainly by filling the stomach and not by stimulating the release of CCK, because consumption of the soup did not increase CCK levels in men, as measured by a 0899-9007/03/$30.00 doi:10.1016/S0899-9007(03)00039-X
554
Nolan et al.
Nutrition Volume 19, Number 6, 2003 TABLE I. NUTRIENT COMPOSITION AND PREPARATION OF TEST FOODS
Ingredient Strawberry yogurt shake Yogurt (g) 8% sucrose (w/v) solution (g) Heavy cream (g) Total weight (g) Total energy (kcal) Energy (%) 800 g of tomato soup Tomato soup (g) Total energy (kcal) Energy (%)
Weight/energy
Protein
Carbohydrate
Fat
216.35 72.11 11.54 300.00 312.00
9.53 0.00 0.23 9.76 39.04 12.47
46.70 5.66 0.35 52.71 210.82 67.36
2.86 0.00 3.96 7.01 63.10 20.16
800.0 327.0
6.7 26.7 8.2
60.0 240.0 73.4
6.7 60.3 18.4
radioimmunoassay.16 However, radioimmunoassay might not be as sensitive as a bioassay for detecting endogenously released CCK, which is comprised of several molecular isoforms. Because other molecular isoforms of CCK are at least as potent as CCK-8 in reducing food intake,17 the present study was undertaken to measure the CCK release stimulated by soup, with a bioassay that detects all known bioactive forms of CCK. To maximize the possibility of stimulating CCK release, a large amount of soup (800 g) was used. A secondary reason for conducting the study was to compare CCK release in men and women, because, in another previous study,18 women were more sensitive than men to the satiating effect of a 640-g tomato soup preload when they ate a meal 30 min after the preload. Therefore, we measured CCK levels after fixed amounts of the soup followed 30 min later by a yogurt shake to simulate conditions employed in earlier intake studies in this laboratory.18,19 This information would be useful in the future for developing quantitative hypotheses about the relation between CCK levels and other possible stimuli that would reduce food intake. Yogurt shakes and soups have been used in these types of studies because they are commonly eaten liquid foods whose gastric emptying could be, and in the case of the soup, has been, measured.18 Men and women also were included in the present study because of the importance of generalizing from men to women or possibly finding differences between them in CCK release. Many previous studies have been done in men, and it is now mandated by the National Institutes of Health that U.S. human studies include both sexes unless there are compelling reasons for the exclusion of one sex.
MATERIALS AND METHODS
Sodium
2.433
juice (Red Cheek) and a standardized lunch consisting of tomato soup (nutrient information is listed in Table I; lot no. CX01AD 1919, expiration date March 1997; Campbell Soup Company, Camden, NJ, USA) and a strawberry yogurt shake20 (Table I). On the first day, subjects consumed 500 g of tomato soup as part of an adaptation trial, and on the test day they consumed 800 g. This quantity was chosen because, in a prior study, 800 g was the average amount eaten when subjects were instructed to eat as much as they could.21 On both days, subjects consumed 300 g of a strawberry yogurt shake 30 min after they began consuming the tomato soup. The quantity of yogurt presented in this study was set to the minimum amount consumed by subjects after consumption of a 480-g tomato soup preload in a prior study.18 The elapsed time of 30 min between the two foods (as preload and test meals) was chosen because it has been used in previous studies in our laboratory. This choice of interval was somewhat arbitrary, but the idea was to allow time for activation of some of the intestinal signals that control food intake. All foods were served in a 32-oz opaque cup. Daily Procedure Each participant reported to the laboratory on 2 non-consecutive days, after an overnight fast, for the breakfast, which was followed 2.5 h later by the soup and yogurt as lunch, as described above. When the subject returned for lunch, an intravenous catheter was inserted into a forearm vein for repeated blood sampling. Isotonic saline (0.9% NaCl) was infused very slowly through the catheter to prevent clotting between sample collections. After 10 min of adaptation to the catheter placement, blood sampling began. Three baseline blood samples were collected at 10-min intervals starting 20 min before consumption of tomato soup. Samples were taken 5,
Subjects The experimental protocol was approved by the Institutional Review Board of St. Luke’s–Roosevelt Hospital Center. Eight nonsmoking, non-obese men (n ⫽ 4) and women (n ⫽ 4) were studied. Each was examined by a physician to ensure that they were in good health. No subject was medicated, and none reported foodrelated allergies or disorders. Their mean age was 23.5 y (range, 18 –28 y), and mean body mass index (weight/height2) was 21.8 kg/m2 (range, 20.5–23.8 kg/m2). We originally planned to do this study with eight people as a pilot, but we obtained a large positive response that we felt was important to publicize. Test Foods The test foods were a standardized 300-kcal breakfast consisting of one Thomas’ English muffin with 7 g of butter and 249 g of apple
FIG. 1. Time line for the sampling of blood, administration of questionnaires (Q), and consumption of food.
Nutrition Volume 19, Number 6, 2003 10, 20, and 30 min after consumption of each food (for food and blood sampling sequence, see Fig. 1). Participants were instructed to drink soup and yogurt shake “as quickly as possible comfortably.” They were timed to ensure that they completed soup ingestion in less than 10 min and the yogurt in less than 5 min. Thirty minutes after subjects began eating the soup, they ate the yogurt shake. Before and after participants consumed the soup and the shake, they completed a questionnaire, as described below. Blood Processing The 11 blood samples (7 mL/sample) were collected in chilled, heparinized vials. They were immediately centrifuged at 1000g for 10 min at 4°C, and the plasma was suction-filtered onto cartridges (Sep-Pak C18, Waters, Milford, MA, USA) and stored at ⫺20°C. Plasma CCK concentrations were measured by a sensitive and specific bioassay that measures all CCK isoforms.6,22
Questionnaires Before and after the soup and the yogurt, subjects were asked to respond to the following questions by marking their ratings on 150-mm line scales, similar to those used by Silverstone and colleagues23: How hungry are you? How strong is your desire for eating your favorite food right now? How full do you feel now? How sick do you feel? How strongly do you feel you have had enough to eat? How much food would you like to eat right now? and How thirsty are you? The lines were anchored at the ends by nothing/none at all (score, 0 mm) and extremely/most I can imagine (score, 150 mm) for feelings/intake.
Data Analysis Each individual’s post-soup and post-yogurt CCK average concentrations (mean of four time points for each), with mean baseline concentration subtracted, were calculated. Areas under the CCK concentration curve exceeding baseline, in the 30 min after the soup and in the 30 min after the yogurt were determined by multiplying the mean level by the time interval and subtracting the baseline values (from before any consumption had taken place)24 The effects of the soup and the yogurt on CCK release were analyzed in two ways. First, paired t tests were performed to determine whether the increases in mean CCK concentration (averaged over the four post-consumption periods) above baseline were significant, and then non-paired t tests were performed separately on the following variables to determine whether there were sex differences in mean CCK concentration and CCK area under the curve after the soup and the shake, the peak CCK level, and the time to reach peak. Second, a two-way (time ⫻ sex) repeated measures analysis of variance was performed on CCK concentrations and on responses to the questionnaire, followed by least significant difference comparisons between sexes for appropriate time points by means of the pooled (within and between groups) error term from the analysis of variance. For CCK concentrations, the contrasts were performed at five time points (times 4 to 8) and the mean of times 4 and 8, which were 5 min after soup and yogurt, respectively. Questionnaire comparisons were done between sexes at four time points, i.e., before and after each food was consumed. Bonferroni’s adjustment for multiple (six for the concentrations and four for the questions) comparisons was done to reduce the possibility that significant results were obtained simply because of multiple testing. The SAS statistical software for the PC was used.25 P ⫽ 0.05 was considered statistically significant for individual tests, and this level would be achieved with P ⫽ 0.0083 for six comparisons and with P ⫽ 0.0125 for four.
Appetitive Ratings and CCK Release After Foods
555
TABLE II. POSTINGESTION CCK INCREASES ABOVE BASELINE IN MEN AND WOMEN (N ⫽ 8)*
Mean SE t P
Post-soup increase over baseline
Post-yogurt increase over baseline
3.813 1.212 3.146 0.016
4.147 1.354 3.064 0.018
Mean (of 5, 10, 20, and 30 min) increases above baseline of plasma CCK concentration after soup and after yogurt shake consumption (pmol/L). CCK, cholecystokinin; SE, standard error.
RESULTS Plasma CCK In the 30 min after tomato soup consumption, mean CCK concentration rose significantly over baseline and remained elevated after yogurt consumption (Table II). There was a significant increase in the area under the curve (over baseline) that was sustained after the yogurt shake (Table III) for both sexes combined. There was no statistically significant difference in mean concentration, the time to reach peak CCK level, or the peak level between men and women. When CCK concentrations were compared between men and women at each time point, the increase over baseline in CCK concentration at 5 min post-soup ingestion was significantly higher for women (7.4 pmol/L ⫾ 2.02 standard error) than for men (0.537 pmol/L ⫾ 0.54) by 6.85 pmol/L (⫾2.81, t ⫽ 2.43, P ⫽ 0.0189; Fig. 2). However, with Bonferroni’s correction (dividing P by 6 for the five post-soup time points and the combined mean of the 5-min points after soup and yogurt), the significance level equivalent to 0.05 is 0.0083; consequently, this difference would not reach statistical significance. However, when CCK concentrations at 5 min after soup and 5 min after yogurt were averaged, the women’s mean averaged concentration was 5.58 pmol/L (⫾ 1.994, t ⫽ 2.80, P ⫽ 0.0073) higher than the men’s, and statistical significance was achieved even with Bonferroni’s adjustment. Analysis of variance also showed a marginally significant interaction (P ⫽ 0.0536) between sex and CCK concentration difference between 5 and 20 min post-soup ingestion. Between these time points, plasma CCK concentrations peaked in men but declined in women (Fig. 2).
TABLE III. AUC FOR CCK RELEASE 30 MIN AFTER SOUP AND AFTER YOGURT SHAKE CONSUMPTION (pmol/L 䡠 30 min) IN MEN AND WOMEN (N ⫽ 8)
AUC (mean) SE t P
Post-soup increase over baseline
post-yogurt increase over baseline
69.19 28.31 2.44 0.040
80.72 28.15 2.87 0.020
AUC, area under the curve; CCK, cholecystokinin; SE, standard error
556
Nolan et al.
Nutrition Volume 19, Number 6, 2003 change significantly (over the four time periods, ratings ranged from 61 mm before soup to 31 mm after yogurt; least significance difference, 33.7 mm). None of the patterns of changes in ratings matched the pattern of changes in CCK levels. There were no sex differences in these ratings. There were sex differences in ratings for the feeling of how sick do you feel (“sick” in Fig. 3, lower left). Women reported significantly higher ratings than did men after soup (71.75 versus 3.75 mm, t6 ⫽ 4.33, P ⫽ 0.0002) and before yogurt: (54.00 versus 3.75 mm, t6 ⫽ 3.20, P ⫽ 0.0039). However, no subjects verbally informed the experimenter that they were feeling at all sick at any time during the study, refused to consume all of their soup, or failed to return for their next appointment because they did not want to consume the soup. The sick and thirsty ratings remained similar across time, unlike the appetite-related ratings and the CCK levels.
FIG. 2. Mean plasma CCK concentration over time. Arrows indicate the times at which tomato soup and strawberry yogurt shake were consumed by men (open symbols) and women (solid symbols). CCK, cholecystokinin.
Questionnaires Ratings of hungry, how much food would you like to eat, and desire for favorite food were mirror images of full, and all changed significantly after consumption of the soup (Fig. 3). Ratings of full rose significantly, from 34 mm before to 123 mm after soup (least significance difference, 22.7), and then dropped to 91 mm before the yogurt and increased to 125 mm after yogurt. Thirsty did not
DISCUSSION In contrast to previous negative results16 obtained with an radioimmunoassay that measured only CCK-8, the present study, obtained with a bioassay that measured all CCK isoforms, showed a significant release of CCK after the ingestion of tomato soup. Consequently, the previous hypothesis, that soup enhances the effectiveness of CCK in reducing food intake exclusively by providing a gastric distending signal that is amplified by CCK, must be tempered with the likelihood that some of the increased effectiveness could have been related to CCK release by the soup. We cannot be certain that endogenously released CCK participated in the reduction of food intake after the 500-g soup preload in the previous study,3 because a larger amount of soup was used in the present, than in the previous, study. However, it
FIG. 3. Selected questionnaire responses at each time point. *P ⬍ 0.05 between males and females. All responses changed significantly after the soup, but did not change significantly after that (i.e., before or after yogurt), except for full, which dropped significantly before yogurt and rose significantly after it.
Nutrition Volume 19, Number 6, 2003 is not known whether there is a dose–response relation between soup consumption and CCK release, or what the threshold is for soup (or other nutrient containing foods) to release CCK. Further, because larger molecular forms of CCK differ in potency and duration of action from CCK-8, it is important that any study of the satiating effects of endogenously released CCK include measurement of these larger molecular forms. These results have implications for other studies in which these foods have been used to understand mechanisms that control food intake in humans. The shake and the soup are food stimuli that have been used in many studies in our laboratory20,26,27 and in others’.28 –30 The fact that these foods elevate CCK levels means that these foods cannot be considered solely as gastric distending stimuli (in the case of soup) or as pure measures of intake effects (in the case of yogurt). Consequently, their consumption provides a source of stimulation that interacts with whatever other manipulation one is trying to study. Further, CCK is not the only hormone whose levels are raised by food consumption, and a complete understanding of the hormonal milieu contributing to the inhibition of eating will require simultaneous measurement of many hormones. The implications beyond the present study are that the effects obtained with test meals used to measure effects of manipulations that control food intake may differ depending on how much CCK the meal releases. However to measure these influences, it will be necessary to determine the dose–response effect of endogenous CCK release on intake of, or desire to eat, foods with different CCK-releasing potential coupled with CCK antagonists, to separate other nutrient-induced, food intake-reducing effects from those specifically attributable to CCK. Of particular interest is the initially faster rise in CCK after soup and yogurt in women than in men. Given the fact that neither the size of the difference nor its time course could be predicted in advance, the current observations must be regarded as tentative, but they clearly suggest that further studies be done. However, because the difference was short-lived, the likelihood that it contributed to differences in responses to preloads between men and women18 is reduced. Sex differences in CCK response should be explored further, particularly because interactions between estrogens and CCK infusions on food intake reduction have been demonstrated in rats.31,32 These results also have implications for studies of the effects of preloads and meals that reach the cognitive level. Men and women reported similar increases in fullness and decreases in hunger. The only significant difference in ratings between the sexes was in the elevated rating for sick in women compared with that in men. It is unlikely that these elevated sick ratings were mediated by the endogenous rise in CCK, because there was no relation between sick ratings and CCK release. The sick rating decreased from before to after the yogurt shake, but the CCK level remained elevated. It is more likely that the sick ratings differed because of a difference in men’s and women’s sensitivities to gastric distending effects of the soup. In conclusion, large tomato soup preloads can stimulate CCK release, but further work is needed to determine whether this effect occurs with smaller amounts of soup, and whether this effect contributed to previous findings in which tomato soup preloads augmented the satiating potency of exogenous CCK-8.
ACKNOWLEDGMENTS The authors thank Yehudit Schacter, Timothy Wentzlaff, and Julie Caretta for their technical assistance in completing the study.
REFERENCES 1. Smith GP, Gibbs J. Satiating effect of cholecystokinin. Ann NY Acad Sci 1994;713:236 2. Moran TH, McHugh PR. Cholecystokinin suppresses food intake by inhibiting gastric emptying. Am J Physiol 1982;242:R491
Appetitive Ratings and CCK Release After Foods
557
3. Muurahainen NE, Kissileff HR, Lachaussee J, Pi-Sunyer FX. Effect of a soup preload on reduction of food intake by cholecystokinin in humans. Am J Physiol 1991;260:R272 4. Byrnes DJ, Henderson L, Borody T, Rehfeld JF. Radioimmunoassay of cholecystokinin in human plasma. Clin Chim Acta 1981;111:81 5. Becker HD, Werner M, Schafmayer A. Release of radioimmunologic cholecystokinin in humans. Am J Surg 1984;147:124 6. Liddle RA, Goldfine ID, Williams JA. Bioassay of plasma cholecystokinin in rats: effects of food, trypsin inhibitor, and alcohol. Gastroenterology 1984;87:542 7. Holt S, Brand J, Soveny C, Hansky J. Relationship of satiety to postprandial glycaemic, insulin, and cholecystokinin responses. Appetite 1992;18:129 8. Hopman WPM, Jansen JBMJ, Lamers CBHW. Comparative study of the effects of equal amounts of fat, protein, and starch on plasma cholecystokinin in man. Scand J Gastroenterol 1985;20:843 9. Friedman MI. Metabolic control of calorie intake. In: Friedman MI, Tordoff MG, Kare MR, eds. Appetite and nutrition. Chemical senses. Vol 4. New York: Marcel Dekker, 1991:19 10. Mossner J, Grumann M, Zeeh J, Fischbach W. Influence of various nutrients and their mode of application on plasma cholecystokinin (CCK) bioactivity. Clin Invest 1992;70:125 11. Lieverse RJ, Jansen JBMJ, Masclee AAM, Lamers CBHW. Role of cholecystokinin in the regulation of satiation and satiety in humans. Ann NY Acad Sci 1994;713:268 12. Wolkowitz OM, Gertz B, Weingartner H, et al. Hunger in humans induced by MK-329, a specific peripheral-type cholecystokinin receptor antagonist. Biol Psychiatry 1990;28:169 13. Gutzwiller JP, Drewe J, Ketterer S, et al. Interaction between CCK and a preload on reduction of food intake is mediated by CCK-A receptors in humans. Am J Physiol Regul Integr Comp Physiol 2000;279:R189 14. French SJ, Bergin A, Sepple CP, Read NW, Rovati L. The effects of loxiglumide on food intake in normal weight volunteers. Int J Obes Relat Metab Disord 1994;18:738 15. Beglinger C, Degen L, Matzinger D, D’Amato M, Drewe J. Loxiglumide, a CCK-A receptor antagonist, stimulates calorie intake, and hunger feelings in humans. Am J Physiol Regul Integr Comp Physiol 2001;280:R1149 16. Muurahainen NE. Effects of bombesin, and cholecystokinin on human food intake (doctoral dissertation). New York: Columbia University, 1988 17. Smith GP, Gibbs G. The satiating effects of cholecystokinin and bombesin-like peptides. In: Smith G, ed. Satiation: from gut to brain. New York: Oxford University Press, 1998:97 18. Kissileff HR, Wentzlaff T, Pierson RN, Pi-Sunyer FX. The relationship between amount of soup in the stomach to the control of test meal intake in men and women. Soc Neurosci Abstr 1994;20:1680 19. Carretta JC, Kissileff HR, Pi-Sunyer FX, Oberman L, Geliebter A. Gastric distention and CCK combine to reduce food intake in humans. Soc Neurosci Abstr 1995;21:10.3 20. Kissileff HR, Walsh T, Kral J, Cassidy S. Laboratory studies of eating behavior in women with bulimia. Physiol Behav 1986;38:563 21. Wentzlaff T, Guss JL, Kissileff HR. Subjective ratings as a function of amount consumed: a preliminary report. Physiol Behav 1995;57:1209 22. Liddle RA, Goldfine ID, Rosen MS, Taplitz RA, Williams JA. Cholecystokinin bioactivity in human plasma: molecular forms, responses to feeding, and relationship to gallbladder contraction. J Clin Invest 1985;75:1144 23. Silverstone T, Goodall E, Brambilla F. Measurement of hunger and food intake. In: Ferrari E, ed. Disorders of eating behavior a psychoneuroendocrine approach. Oxford: Pergamon Press, 1986:129 24. Tai MM. A mathematical model for the determination of total area under glucose tolerance and other metabolic curves. Diabet Care 1994;17:152 25. SAS/STAT user’s guide, release 6.03. Cary: SAS Institute, 1988 26. Kissileff HR, Pi-Sunyer FX, Segal K, Meltzer S, Foelsch PA. Acute effects of exercise on food intake in obese and nonobese women. Am J Clin Nutr 1990; 52:240 27. Kissileff HR, Gruss LP, Thornton J, Jordan HA. The satiating efficiency of foods. Physiol Behav 1984;32:319 28. Rolls BJ, Kim-Harris S, Fischman MW, et al. Satiety after preloads with different amounts of fat and carbohydrate: implications for obesity. Am J Clin Nutr 1994;60:476 29. Rolls BJ, Fedoroff IC, Guthrie JF, Laster LJ. Foods with different satiating effects in humans. Appetite 1990;15:115 30. Hill AJ, Peikin SR, Ryan CA, Blundell JE. Oral administration of proteinase inhibitor II from potatoes reduces energy intake in man. Physiol Behav 1990;48: 241 31. Geary N, Trace D, McEwen B, Smith GP. Cyclic estradiol replacement increases the satiety effect of CCK-8 in ovariectomized rats. Physiol Behav 1994;56:281 32. Geary N. Estradiol, CCK, and satiation. Peptides 2001;22:1251
Elevated Plasma Cholecystokinin and Appetitive Ratings After Consumption of a Liquid Meal in Humans Laurence J. Nolan, PhD, Janet L. Guss, MS, Rodger A. Liddle, MD, F. Xavier Pi-Sunyer, MD, and Harry R. Kissileff, PhD From the New York Obesity Research Center, St. Luke’s–Roosevelt Hospital Center and Columbia University College of Physicians & Surgeons, New York, New York, USA; and the Department of Medicine, Duke University Medical Center, Durham, North Carolina, USA OBJECTIVE: This study had two objectives. The first was to evaluate the possibility that, in a previous study, a soup preload augmented the reduction of food intake in a test meal induced by an exogenous infusion of cholecystokinin (CCK) because the soup also endogenously released CCK. The second was to compare CCK release by soup between men and women to determine whether the increased satiating effectiveness of soup in women as opposed to men could have been partly attributable to differences in CCK release. METHODS: By using a bioassay that measures all of its known isoforms, we determined plasma CCK levels at baseline and at several times postprandially in eight healthy, non-obese men and women (four of each sex). Each subject ingested 800 g of tomato soup, which was followed 30 min later by 300 g of a yogurt shake. Appetitive ratings were also collected and related to CCK levels. RESULTS: Ingestion of tomato soup significantly increased plasma CCK levels by 3.81 pmol/L (⫾ 1.21 standard error, P ⫽ 0.016) over baseline within 30 min in all subjects combined. When CCK concentrations at 5 min after soup and 5 min after yogurt were averaged, the women’s mean averaged concentration was 5.58 pmol/L (⫾ 1.994, t ⫽ 2.80, P ⫽ 0.0073) higher than the men’s. The elevated levels persisted but did not rise further upon consumption of the yogurt shake. Hunger ratings declined and fullness ratings increased after eating, although patterns of ratings did not match exactly patterns of CCK release. CONCLUSIONS: A large quantity of tomato soup stimulates significant CCK release; therefore, some of the satiating effects of soup preloads could have been mediated by an elevation in endogenous CCK. Nutrition 2003;19:553–557. ©Elsevier Inc. 2003 KEY WORDS: cholecystokinin, human ingestive behavior, satiety, eating behavior
INTRODUCTION The influence of cholecystokinin (CCK) in the control of food intake has been investigated for almost 30 y, but its role in normal eating episodes in humans has yet to be fully explored. The evidence for activity of CCK in the control of food intake includes the following facts. The exogenous administration of CCK reduces food intake in all species tested, including humans (for review, see Smith and Gibbs1), and CCK antagonists increase food intake in animals (for review, see Smith and Gibbs1). After the demonstra-
This study was supported in part by the following grants from the National Institutes of Health: DK36507, DK53089, and T32 DK07559 to St. Luke’s/ Roosevelt Institute for Health Sciences (H. R. Kissileff, principal investigator); DK38626 to Duke University Medical Center (R. A. Liddle, principal investigator.); and DK26687 to The New York Obesity Research Center (F. X. Pi-Sunyer, Director). Laurence J. Nolan’s current address: Department of Psychology, Wagner College, Staten Island, NY, 10301, USA Correspondence to: Harry R. Kissileff, MD, St. Luke’s-Roosevelt Hospital Center, 1111 Amsterdam Avenue, New York, NY 10025, USA. E-mail: [email protected] Nutrition 19:553–557, 2003 ©Elsevier Inc., 2003. Printed in the United States. All rights reserved.
tion by Moran and McHugh,2 that a saline preload is necessary for a low dose of CCK to reduce food intake in monkeys, our laboratory showed that reduction in food intake induced by exogenous administration of CCK in humans is enhanced significantly by combining it with a preload of soup.3 Post-meal release of CCK has been demonstrated in humans4 – 6 and correlated with verbal ratings of satiety.7 Proteins and fats are the best macronutrient secretagogues for CCK in humans.8 –10 Antagonists specific for the CCK-A receptor have been shown to reduce the satiating potency of intestinal fat infusions (loxiglumide)11 and increase hunger ratings (devazepide).12 Further, loxiglumide blocked the effect of CCK to suppress food intake when combined with preload of bananas and water.13 Conversely, French et al.14 reported that loxiglumide has no effect on hunger ratings. In three studies after administration of loxiglumide, food intake (in units of weight) in humans11,14,15 did not increase significantly, although energy intake increased significantly by 10% in one study.15 Thus, these studies showed a role for endogenous CCK in sensations of satiety in humans, although its magnitude and the essential conditions for its effectiveness remain elusive. The results of a previous study3 suggested that a nutrient-dilute liquid food such as tomato soup augments satiety induced by an intravenous CCK infusion, mainly by filling the stomach and not by stimulating the release of CCK, because consumption of the soup did not increase CCK levels in men, as measured by a 0899-9007/03/$30.00 doi:10.1016/S0899-9007(03)00039-X
554
Nolan et al.
Nutrition Volume 19, Number 6, 2003 TABLE I. NUTRIENT COMPOSITION AND PREPARATION OF TEST FOODS
Ingredient Strawberry yogurt shake Yogurt (g) 8% sucrose (w/v) solution (g) Heavy cream (g) Total weight (g) Total energy (kcal) Energy (%) 800 g of tomato soup Tomato soup (g) Total energy (kcal) Energy (%)
Weight/energy
Protein
Carbohydrate
Fat
216.35 72.11 11.54 300.00 312.00
9.53 0.00 0.23 9.76 39.04 12.47
46.70 5.66 0.35 52.71 210.82 67.36
2.86 0.00 3.96 7.01 63.10 20.16
800.0 327.0
6.7 26.7 8.2
60.0 240.0 73.4
6.7 60.3 18.4
radioimmunoassay.16 However, radioimmunoassay might not be as sensitive as a bioassay for detecting endogenously released CCK, which is comprised of several molecular isoforms. Because other molecular isoforms of CCK are at least as potent as CCK-8 in reducing food intake,17 the present study was undertaken to measure the CCK release stimulated by soup, with a bioassay that detects all known bioactive forms of CCK. To maximize the possibility of stimulating CCK release, a large amount of soup (800 g) was used. A secondary reason for conducting the study was to compare CCK release in men and women, because, in another previous study,18 women were more sensitive than men to the satiating effect of a 640-g tomato soup preload when they ate a meal 30 min after the preload. Therefore, we measured CCK levels after fixed amounts of the soup followed 30 min later by a yogurt shake to simulate conditions employed in earlier intake studies in this laboratory.18,19 This information would be useful in the future for developing quantitative hypotheses about the relation between CCK levels and other possible stimuli that would reduce food intake. Yogurt shakes and soups have been used in these types of studies because they are commonly eaten liquid foods whose gastric emptying could be, and in the case of the soup, has been, measured.18 Men and women also were included in the present study because of the importance of generalizing from men to women or possibly finding differences between them in CCK release. Many previous studies have been done in men, and it is now mandated by the National Institutes of Health that U.S. human studies include both sexes unless there are compelling reasons for the exclusion of one sex.
MATERIALS AND METHODS
Sodium
2.433
juice (Red Cheek) and a standardized lunch consisting of tomato soup (nutrient information is listed in Table I; lot no. CX01AD 1919, expiration date March 1997; Campbell Soup Company, Camden, NJ, USA) and a strawberry yogurt shake20 (Table I). On the first day, subjects consumed 500 g of tomato soup as part of an adaptation trial, and on the test day they consumed 800 g. This quantity was chosen because, in a prior study, 800 g was the average amount eaten when subjects were instructed to eat as much as they could.21 On both days, subjects consumed 300 g of a strawberry yogurt shake 30 min after they began consuming the tomato soup. The quantity of yogurt presented in this study was set to the minimum amount consumed by subjects after consumption of a 480-g tomato soup preload in a prior study.18 The elapsed time of 30 min between the two foods (as preload and test meals) was chosen because it has been used in previous studies in our laboratory. This choice of interval was somewhat arbitrary, but the idea was to allow time for activation of some of the intestinal signals that control food intake. All foods were served in a 32-oz opaque cup. Daily Procedure Each participant reported to the laboratory on 2 non-consecutive days, after an overnight fast, for the breakfast, which was followed 2.5 h later by the soup and yogurt as lunch, as described above. When the subject returned for lunch, an intravenous catheter was inserted into a forearm vein for repeated blood sampling. Isotonic saline (0.9% NaCl) was infused very slowly through the catheter to prevent clotting between sample collections. After 10 min of adaptation to the catheter placement, blood sampling began. Three baseline blood samples were collected at 10-min intervals starting 20 min before consumption of tomato soup. Samples were taken 5,
Subjects The experimental protocol was approved by the Institutional Review Board of St. Luke’s–Roosevelt Hospital Center. Eight nonsmoking, non-obese men (n ⫽ 4) and women (n ⫽ 4) were studied. Each was examined by a physician to ensure that they were in good health. No subject was medicated, and none reported foodrelated allergies or disorders. Their mean age was 23.5 y (range, 18 –28 y), and mean body mass index (weight/height2) was 21.8 kg/m2 (range, 20.5–23.8 kg/m2). We originally planned to do this study with eight people as a pilot, but we obtained a large positive response that we felt was important to publicize. Test Foods The test foods were a standardized 300-kcal breakfast consisting of one Thomas’ English muffin with 7 g of butter and 249 g of apple
FIG. 1. Time line for the sampling of blood, administration of questionnaires (Q), and consumption of food.
Nutrition Volume 19, Number 6, 2003 10, 20, and 30 min after consumption of each food (for food and blood sampling sequence, see Fig. 1). Participants were instructed to drink soup and yogurt shake “as quickly as possible comfortably.” They were timed to ensure that they completed soup ingestion in less than 10 min and the yogurt in less than 5 min. Thirty minutes after subjects began eating the soup, they ate the yogurt shake. Before and after participants consumed the soup and the shake, they completed a questionnaire, as described below. Blood Processing The 11 blood samples (7 mL/sample) were collected in chilled, heparinized vials. They were immediately centrifuged at 1000g for 10 min at 4°C, and the plasma was suction-filtered onto cartridges (Sep-Pak C18, Waters, Milford, MA, USA) and stored at ⫺20°C. Plasma CCK concentrations were measured by a sensitive and specific bioassay that measures all CCK isoforms.6,22
Questionnaires Before and after the soup and the yogurt, subjects were asked to respond to the following questions by marking their ratings on 150-mm line scales, similar to those used by Silverstone and colleagues23: How hungry are you? How strong is your desire for eating your favorite food right now? How full do you feel now? How sick do you feel? How strongly do you feel you have had enough to eat? How much food would you like to eat right now? and How thirsty are you? The lines were anchored at the ends by nothing/none at all (score, 0 mm) and extremely/most I can imagine (score, 150 mm) for feelings/intake.
Data Analysis Each individual’s post-soup and post-yogurt CCK average concentrations (mean of four time points for each), with mean baseline concentration subtracted, were calculated. Areas under the CCK concentration curve exceeding baseline, in the 30 min after the soup and in the 30 min after the yogurt were determined by multiplying the mean level by the time interval and subtracting the baseline values (from before any consumption had taken place)24 The effects of the soup and the yogurt on CCK release were analyzed in two ways. First, paired t tests were performed to determine whether the increases in mean CCK concentration (averaged over the four post-consumption periods) above baseline were significant, and then non-paired t tests were performed separately on the following variables to determine whether there were sex differences in mean CCK concentration and CCK area under the curve after the soup and the shake, the peak CCK level, and the time to reach peak. Second, a two-way (time ⫻ sex) repeated measures analysis of variance was performed on CCK concentrations and on responses to the questionnaire, followed by least significant difference comparisons between sexes for appropriate time points by means of the pooled (within and between groups) error term from the analysis of variance. For CCK concentrations, the contrasts were performed at five time points (times 4 to 8) and the mean of times 4 and 8, which were 5 min after soup and yogurt, respectively. Questionnaire comparisons were done between sexes at four time points, i.e., before and after each food was consumed. Bonferroni’s adjustment for multiple (six for the concentrations and four for the questions) comparisons was done to reduce the possibility that significant results were obtained simply because of multiple testing. The SAS statistical software for the PC was used.25 P ⫽ 0.05 was considered statistically significant for individual tests, and this level would be achieved with P ⫽ 0.0083 for six comparisons and with P ⫽ 0.0125 for four.
Appetitive Ratings and CCK Release After Foods
555
TABLE II. POSTINGESTION CCK INCREASES ABOVE BASELINE IN MEN AND WOMEN (N ⫽ 8)*
Mean SE t P
Post-soup increase over baseline
Post-yogurt increase over baseline
3.813 1.212 3.146 0.016
4.147 1.354 3.064 0.018
Mean (of 5, 10, 20, and 30 min) increases above baseline of plasma CCK concentration after soup and after yogurt shake consumption (pmol/L). CCK, cholecystokinin; SE, standard error.
RESULTS Plasma CCK In the 30 min after tomato soup consumption, mean CCK concentration rose significantly over baseline and remained elevated after yogurt consumption (Table II). There was a significant increase in the area under the curve (over baseline) that was sustained after the yogurt shake (Table III) for both sexes combined. There was no statistically significant difference in mean concentration, the time to reach peak CCK level, or the peak level between men and women. When CCK concentrations were compared between men and women at each time point, the increase over baseline in CCK concentration at 5 min post-soup ingestion was significantly higher for women (7.4 pmol/L ⫾ 2.02 standard error) than for men (0.537 pmol/L ⫾ 0.54) by 6.85 pmol/L (⫾2.81, t ⫽ 2.43, P ⫽ 0.0189; Fig. 2). However, with Bonferroni’s correction (dividing P by 6 for the five post-soup time points and the combined mean of the 5-min points after soup and yogurt), the significance level equivalent to 0.05 is 0.0083; consequently, this difference would not reach statistical significance. However, when CCK concentrations at 5 min after soup and 5 min after yogurt were averaged, the women’s mean averaged concentration was 5.58 pmol/L (⫾ 1.994, t ⫽ 2.80, P ⫽ 0.0073) higher than the men’s, and statistical significance was achieved even with Bonferroni’s adjustment. Analysis of variance also showed a marginally significant interaction (P ⫽ 0.0536) between sex and CCK concentration difference between 5 and 20 min post-soup ingestion. Between these time points, plasma CCK concentrations peaked in men but declined in women (Fig. 2).
TABLE III. AUC FOR CCK RELEASE 30 MIN AFTER SOUP AND AFTER YOGURT SHAKE CONSUMPTION (pmol/L 䡠 30 min) IN MEN AND WOMEN (N ⫽ 8)
AUC (mean) SE t P
Post-soup increase over baseline
post-yogurt increase over baseline
69.19 28.31 2.44 0.040
80.72 28.15 2.87 0.020
AUC, area under the curve; CCK, cholecystokinin; SE, standard error
556
Nolan et al.
Nutrition Volume 19, Number 6, 2003 change significantly (over the four time periods, ratings ranged from 61 mm before soup to 31 mm after yogurt; least significance difference, 33.7 mm). None of the patterns of changes in ratings matched the pattern of changes in CCK levels. There were no sex differences in these ratings. There were sex differences in ratings for the feeling of how sick do you feel (“sick” in Fig. 3, lower left). Women reported significantly higher ratings than did men after soup (71.75 versus 3.75 mm, t6 ⫽ 4.33, P ⫽ 0.0002) and before yogurt: (54.00 versus 3.75 mm, t6 ⫽ 3.20, P ⫽ 0.0039). However, no subjects verbally informed the experimenter that they were feeling at all sick at any time during the study, refused to consume all of their soup, or failed to return for their next appointment because they did not want to consume the soup. The sick and thirsty ratings remained similar across time, unlike the appetite-related ratings and the CCK levels.
FIG. 2. Mean plasma CCK concentration over time. Arrows indicate the times at which tomato soup and strawberry yogurt shake were consumed by men (open symbols) and women (solid symbols). CCK, cholecystokinin.
Questionnaires Ratings of hungry, how much food would you like to eat, and desire for favorite food were mirror images of full, and all changed significantly after consumption of the soup (Fig. 3). Ratings of full rose significantly, from 34 mm before to 123 mm after soup (least significance difference, 22.7), and then dropped to 91 mm before the yogurt and increased to 125 mm after yogurt. Thirsty did not
DISCUSSION In contrast to previous negative results16 obtained with an radioimmunoassay that measured only CCK-8, the present study, obtained with a bioassay that measured all CCK isoforms, showed a significant release of CCK after the ingestion of tomato soup. Consequently, the previous hypothesis, that soup enhances the effectiveness of CCK in reducing food intake exclusively by providing a gastric distending signal that is amplified by CCK, must be tempered with the likelihood that some of the increased effectiveness could have been related to CCK release by the soup. We cannot be certain that endogenously released CCK participated in the reduction of food intake after the 500-g soup preload in the previous study,3 because a larger amount of soup was used in the present, than in the previous, study. However, it
FIG. 3. Selected questionnaire responses at each time point. *P ⬍ 0.05 between males and females. All responses changed significantly after the soup, but did not change significantly after that (i.e., before or after yogurt), except for full, which dropped significantly before yogurt and rose significantly after it.
Nutrition Volume 19, Number 6, 2003 is not known whether there is a dose–response relation between soup consumption and CCK release, or what the threshold is for soup (or other nutrient containing foods) to release CCK. Further, because larger molecular forms of CCK differ in potency and duration of action from CCK-8, it is important that any study of the satiating effects of endogenously released CCK include measurement of these larger molecular forms. These results have implications for other studies in which these foods have been used to understand mechanisms that control food intake in humans. The shake and the soup are food stimuli that have been used in many studies in our laboratory20,26,27 and in others’.28 –30 The fact that these foods elevate CCK levels means that these foods cannot be considered solely as gastric distending stimuli (in the case of soup) or as pure measures of intake effects (in the case of yogurt). Consequently, their consumption provides a source of stimulation that interacts with whatever other manipulation one is trying to study. Further, CCK is not the only hormone whose levels are raised by food consumption, and a complete understanding of the hormonal milieu contributing to the inhibition of eating will require simultaneous measurement of many hormones. The implications beyond the present study are that the effects obtained with test meals used to measure effects of manipulations that control food intake may differ depending on how much CCK the meal releases. However to measure these influences, it will be necessary to determine the dose–response effect of endogenous CCK release on intake of, or desire to eat, foods with different CCK-releasing potential coupled with CCK antagonists, to separate other nutrient-induced, food intake-reducing effects from those specifically attributable to CCK. Of particular interest is the initially faster rise in CCK after soup and yogurt in women than in men. Given the fact that neither the size of the difference nor its time course could be predicted in advance, the current observations must be regarded as tentative, but they clearly suggest that further studies be done. However, because the difference was short-lived, the likelihood that it contributed to differences in responses to preloads between men and women18 is reduced. Sex differences in CCK response should be explored further, particularly because interactions between estrogens and CCK infusions on food intake reduction have been demonstrated in rats.31,32 These results also have implications for studies of the effects of preloads and meals that reach the cognitive level. Men and women reported similar increases in fullness and decreases in hunger. The only significant difference in ratings between the sexes was in the elevated rating for sick in women compared with that in men. It is unlikely that these elevated sick ratings were mediated by the endogenous rise in CCK, because there was no relation between sick ratings and CCK release. The sick rating decreased from before to after the yogurt shake, but the CCK level remained elevated. It is more likely that the sick ratings differed because of a difference in men’s and women’s sensitivities to gastric distending effects of the soup. In conclusion, large tomato soup preloads can stimulate CCK release, but further work is needed to determine whether this effect occurs with smaller amounts of soup, and whether this effect contributed to previous findings in which tomato soup preloads augmented the satiating potency of exogenous CCK-8.
ACKNOWLEDGMENTS The authors thank Yehudit Schacter, Timothy Wentzlaff, and Julie Caretta for their technical assistance in completing the study.
REFERENCES 1. Smith GP, Gibbs J. Satiating effect of cholecystokinin. Ann NY Acad Sci 1994;713:236 2. Moran TH, McHugh PR. Cholecystokinin suppresses food intake by inhibiting gastric emptying. Am J Physiol 1982;242:R491
Appetitive Ratings and CCK Release After Foods
557
3. Muurahainen NE, Kissileff HR, Lachaussee J, Pi-Sunyer FX. Effect of a soup preload on reduction of food intake by cholecystokinin in humans. Am J Physiol 1991;260:R272 4. Byrnes DJ, Henderson L, Borody T, Rehfeld JF. Radioimmunoassay of cholecystokinin in human plasma. Clin Chim Acta 1981;111:81 5. Becker HD, Werner M, Schafmayer A. Release of radioimmunologic cholecystokinin in humans. Am J Surg 1984;147:124 6. Liddle RA, Goldfine ID, Williams JA. Bioassay of plasma cholecystokinin in rats: effects of food, trypsin inhibitor, and alcohol. Gastroenterology 1984;87:542 7. Holt S, Brand J, Soveny C, Hansky J. Relationship of satiety to postprandial glycaemic, insulin, and cholecystokinin responses. Appetite 1992;18:129 8. Hopman WPM, Jansen JBMJ, Lamers CBHW. Comparative study of the effects of equal amounts of fat, protein, and starch on plasma cholecystokinin in man. Scand J Gastroenterol 1985;20:843 9. Friedman MI. Metabolic control of calorie intake. In: Friedman MI, Tordoff MG, Kare MR, eds. Appetite and nutrition. Chemical senses. Vol 4. New York: Marcel Dekker, 1991:19 10. Mossner J, Grumann M, Zeeh J, Fischbach W. Influence of various nutrients and their mode of application on plasma cholecystokinin (CCK) bioactivity. Clin Invest 1992;70:125 11. Lieverse RJ, Jansen JBMJ, Masclee AAM, Lamers CBHW. Role of cholecystokinin in the regulation of satiation and satiety in humans. Ann NY Acad Sci 1994;713:268 12. Wolkowitz OM, Gertz B, Weingartner H, et al. Hunger in humans induced by MK-329, a specific peripheral-type cholecystokinin receptor antagonist. Biol Psychiatry 1990;28:169 13. Gutzwiller JP, Drewe J, Ketterer S, et al. Interaction between CCK and a preload on reduction of food intake is mediated by CCK-A receptors in humans. Am J Physiol Regul Integr Comp Physiol 2000;279:R189 14. French SJ, Bergin A, Sepple CP, Read NW, Rovati L. The effects of loxiglumide on food intake in normal weight volunteers. Int J Obes Relat Metab Disord 1994;18:738 15. Beglinger C, Degen L, Matzinger D, D’Amato M, Drewe J. Loxiglumide, a CCK-A receptor antagonist, stimulates calorie intake, and hunger feelings in humans. Am J Physiol Regul Integr Comp Physiol 2001;280:R1149 16. Muurahainen NE. Effects of bombesin, and cholecystokinin on human food intake (doctoral dissertation). New York: Columbia University, 1988 17. Smith GP, Gibbs G. The satiating effects of cholecystokinin and bombesin-like peptides. In: Smith G, ed. Satiation: from gut to brain. New York: Oxford University Press, 1998:97 18. Kissileff HR, Wentzlaff T, Pierson RN, Pi-Sunyer FX. The relationship between amount of soup in the stomach to the control of test meal intake in men and women. Soc Neurosci Abstr 1994;20:1680 19. Carretta JC, Kissileff HR, Pi-Sunyer FX, Oberman L, Geliebter A. Gastric distention and CCK combine to reduce food intake in humans. Soc Neurosci Abstr 1995;21:10.3 20. Kissileff HR, Walsh T, Kral J, Cassidy S. Laboratory studies of eating behavior in women with bulimia. Physiol Behav 1986;38:563 21. Wentzlaff T, Guss JL, Kissileff HR. Subjective ratings as a function of amount consumed: a preliminary report. Physiol Behav 1995;57:1209 22. Liddle RA, Goldfine ID, Rosen MS, Taplitz RA, Williams JA. Cholecystokinin bioactivity in human plasma: molecular forms, responses to feeding, and relationship to gallbladder contraction. J Clin Invest 1985;75:1144 23. Silverstone T, Goodall E, Brambilla F. Measurement of hunger and food intake. In: Ferrari E, ed. Disorders of eating behavior a psychoneuroendocrine approach. Oxford: Pergamon Press, 1986:129 24. Tai MM. A mathematical model for the determination of total area under glucose tolerance and other metabolic curves. Diabet Care 1994;17:152 25. SAS/STAT user’s guide, release 6.03. Cary: SAS Institute, 1988 26. Kissileff HR, Pi-Sunyer FX, Segal K, Meltzer S, Foelsch PA. Acute effects of exercise on food intake in obese and nonobese women. Am J Clin Nutr 1990; 52:240 27. Kissileff HR, Gruss LP, Thornton J, Jordan HA. The satiating efficiency of foods. Physiol Behav 1984;32:319 28. Rolls BJ, Kim-Harris S, Fischman MW, et al. Satiety after preloads with different amounts of fat and carbohydrate: implications for obesity. Am J Clin Nutr 1994;60:476 29. Rolls BJ, Fedoroff IC, Guthrie JF, Laster LJ. Foods with different satiating effects in humans. Appetite 1990;15:115 30. Hill AJ, Peikin SR, Ryan CA, Blundell JE. Oral administration of proteinase inhibitor II from potatoes reduces energy intake in man. Physiol Behav 1990;48: 241 31. Geary N, Trace D, McEwen B, Smith GP. Cyclic estradiol replacement increases the satiety effect of CCK-8 in ovariectomized rats. Physiol Behav 1994;56:281 32. Geary N. Estradiol, CCK, and satiation. Peptides 2001;22:1251