| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (A.D.S., A.D., C.A.V., P.A.T.), Phoenix, Arizona 85016; and Department of Psychiatry (M.H.T.), Obesity Research Center, University of Cincinnati-Genome Research Institute, Cincinnati, Ohio 45237
Address all correspondence and requests for reprints to: Arline D. Salbe, Ph.D. National Institutes of Health/National Institute of Diabetes and Digestive and Kidney Diseases/Clinical Diabetes and Nutrition Section 4212 North 16th Street, Room 541 Phoenix, AZ 85016. Email: arline_salbe{at}nih.gov
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Despite the mounting evidence that ghrelin acutely stimulates food intake, cross-sectional studies report that endogenous fasting ghrelin concentrations are higher in anorexia nervosa (13) and lower in obese, compared with normal-weight, adults (14) and children (15). To explain this paradox, it has been proposed that plasma ghrelin concentrations rise and fall as an adaptive response to chronic energy deficits and surfeits, respectively (14). This line of reasoning, therefore, would imply that, if ghrelin has an etiologic role in the development of obesity, it should initially be high and drive the hyperphagia that leads to weight gain, before undergoing a compensatory fall to the low levels observed in obese humans. Thus, if endogenous ghrelin concentrations were modulators of food intake, we would expect subjects with relative hyperghrelinemia, i.e. individuals with elevated ghrelin concentrations relative to their body size, to display a greater propensity to overeating than subjects considered relatively hypoghrelinemic. To the best of our knowledge, there are no reports indicating that interindividual differences in plasma ghrelin concentrations within the physiologic range affect ad libitum energy intake in humans.
The current study, therefore, was designed to test the hypothesis that fasting relative hyperghrelinemia is associated with increased energy intake under ad libitum conditions. Because assessing ad libitum food intake in free-living conditions is highly dependent on reporting accuracy (16), we have designed a computer-operated vending machine system to perform that function within the confines of our metabolic unit.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Fifteen Pima Indians (8 male/7 female) were recruited from the Gila River Indian Community, 40 miles southeast of Phoenix. Fifteen Caucasian subjects (12 male/3 female) were recruited from the greater Phoenix area through advertisement. Before participation, volunteers were fully informed of the nature and purpose of the study, and written informed consent was obtained. The experimental protocol was approved by the Institutional Review Board of the National Institute of Diabetes and Digestive and Kidney Diseases and by the Tribal Council of the Gila River Indian Community.
All subjects were found to be free of disease according to physical examination, medical history, and laboratory testing. Upon admission to the metabolic ward, subjects were placed on a standard weight maintenance diet (20, 30, and 50% of daily calories provided as protein, fat, and carbohydrate, respectively) for 3 d before testing. Weight maintenance energy needs (WMEN) on the metabolic ward were calculated for each subject based on weight and gender (men, WMEN = 9.5 x weight (kg) + 1973; women, WMEN = 9.5 x weight (kg) + 1745) (17). Body composition was measured by dual-energy x-ray absorptiometry (DPX-L, Lunar Corp., Madison, WI) as previously described (18). Glucose tolerance was assessed by a 75-g oral glucose tolerance test (19) according to the criteria of the World Health Organization. Only nondiabetic subjects participated in this study.
Assessment of food preferences
After admission to the metabolic ward, subjects were asked to complete a Food Preferences Questionnaire consisting of a listing of 80 food items presented in random order. Based on a model developed by Geiselman et al. (20), typical breakfast, lunch, dinner, and snack food items were categorized according to a macronutrient self-selection paradigm that varied the fat content of foods as a percentage of calories systematically with other macronutrients. Foods were categorized as being high (>45% kcal) or low (<20% kcal) in fat and within each of these categories, high in simple sugar (>30% kcal), complex carbohydrate (>30% kcal), or protein (>13% kcal). In completing this self-administered questionnaire, individuals were asked to assign each food a hedonic rating using a 9-point Likert scale with the following anchors: never tasted, 1 = dislike extremely; 5 = neutral; 9 = like extremely. Several foods on the list are among the top 10 sources of dietary fat in the United States, including hamburgers, French fries, ham and other luncheon meats, doughnuts, cookies, cakes, candies, bread products, muffins, eggs, and cheeses; the list also reflects items of intake common to Pima Indians and individuals living in the Southwest.
Ad libitum food intake using a computerized vending machine system
During the final 3 d of study on the metabolic ward, subjects were asked to self-select all their food using a computer-operated vending machine system as previously described (21). The 40 food items made available to the subjects on each of the 3 d consisted of those foods to which the subject had assigned an intermediate (between 4 and 8) hedonic rating on the Food Preferences Questionnaire. In addition, a core group of condiments was provided to each subject each day, including butter, peanut butter, cream cheese, and jams; salad items and dressings; crackers, bread, tortillas, and Indian fry bread; spices and salsa; and orange juice, apple juice, milk, and a six-pack of soda of the subject’s choice. Subjects had ad libitum access to the vending machines for 231/2 h/d. The refrigerated machines were housed in a separate eating area equipped with a table, chair, microwave oven, and toaster. Subjects were instructed to eat only in the vending room and to eat whatever they wished whenever they desired. Television viewing during food consumption was prohibited. Food wrappers and unconsumed food portions were returned to the vending machines.
Daily energy intake (DEI) and protein, fat, and carbohydrate intakes were calculated from the actual weights of food and condiments consumed using the CBORD Professional Diet Analyzer Program (CBORD, Inc., Ithaca, NY) with the database modified to reflect the nutrient content of specific food items as indicated by the manufacturer. Results are presented as the mean ± SD of the 3 d. DEI is expressed as mean kilocalories per day and mean percentage of WMEN (%WMEN) on the metabolic ward ([mean daily energy consumed/WMEN] x 100). Fat intake is expressed as mean percentage of kilocalories per day.
Eating behavior
The Three-Factor Eating Questionnaire (22), which classifies eating behavior among individuals on the basis of dietary restraint, disinhibition, and perceived hunger, and the Gormally Binge Eating Scale (23), which discriminates among persons having no, or moderate, or severe binge eating problems, were used to assess eating behavior. The questionnaires were self-administered by each subject before the ad libitum use of the vending machines.
Analytical measurements
Blood samples were drawn after an overnight fast, at least 3 full days after subjects had consumed a weight maintenance diet on the metabolic ward and before the ad libitum use of the vending machines. Plasma glucose concentrations were measured using the glucose oxidase method (Beckman Instruments Inc., Fullerton, CA). Plasma insulin concentrations were measured with an automated RIA (Concept 4, ICN Biochemicals, Costa Mesa, CA). Two plasma samples for ghrelin analysis were collected under fasting conditions at the time of the oral glucose tolerance test; results represent the mean of the two fasting samples. Plasma ghrelin concentrations were measured with a commercial RIA (Phoenix Pharmaceuticals Inc., Belmont, CA) that uses 125I-labeled bioactive ghrelin as a tracer molecule and a polyclonal antibody raised in rabbits against full-length, octanoylated human ghrelin as previously described (14).
Data analysis
All statistical analyses were performed using software of the SAS Institute (Cary, NC). Throughout the text, the data are expressed as means ± SD. Fasting plasma insulin and ghrelin concentrations were log transformed (log10) to normalize the distributions before analysis. Race and gender differences were assessed by Student’s t test and linear regression modeling. Relationships between the anthropometric and intake variables and plasma ghrelin concentrations were assessed using Spearman correlation coefficients. Linear regression modeling was used to adjust plasma ghrelin concentrations [for body weight or body mass index (BMI), and race and gender] and dietary intake (for body weight or BMI and gender) and to model the effects of ghrelin on intake parameters.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. There were no significant differences in age, body weight, BMI, or percent body fat between the races. Although the gender distribution between the races was dissimilar, this difference was not significant (P = 0.12, 
2 analysis). There were no significant gender differences in age, body weight, or BMI; however, percent body fat was significantly (P = 0.001) lower in men compared with women.
|
After adjustment of the energy intake variables for body weight and gender, there were no racial differences in energy intake or the percent of calories from fat between Pima Indians and Caucasians (Table 1
). After adjustment of the energy intake variables for body weight and race, however, DEI was significantly higher in men compared with women (P = 0.03); however, fat intake as a percentage of calories tended to be higher in women compared with men (P = 0.08).
Fasting plasma ghrelin concentrations were twice as high in Caucasians as in Pima Indians (103 ± 53 vs. 52 ± 18 fmol/ml, P = 0.01); after adjustment for gender and body weight, ghrelin concentrations remained significantly higher in Caucasians than Pima Indians (body weight, 107 ± 45 vs. 48 ± 15 fmol/ml, respectively; P < 0.0001). After adjustment for race and gender, there was a significant negative correlation between ghrelin concentrations and body weight (r = –0.45, P = 0.01; Fig. 1) as well as BMI (r = –0.50, P = 0.005), but not percent body fat (r = –0.29, P = 0.12). After adjustment for race and body weight, ghrelin concentrations were lower in men compared with women (body weight, 69 ± 29 vs. 94 ± 39 fmol/ml, respectively; P = 0.06).
|
. For any given body weight, there is a range of ghrelin concentrations. In terms of absolute plasma ghrelin concentrations, hypothetical lean person A has a high ghrelin value, whereas hypothetical obese person B has a low ghrelin value. In terms of relative ghrelinemia, both A and B fall on the regression line and may be considered normoghrelinemic for their weight. In contrast, actual subjects C and D have similar BMI values, but subject C is considered relatively hyperghrelinemic, whereas subject D is considered relatively hypoghrelinemic.
Fasting plasma ghrelin concentrations were a significant negative determinant of intake, whether assessed as DEI, %WMEN, or percent of calories as fat (Table 2). And, after adjustment for race, gender, and body weight, relative hyperghrelinemia was negatively correlated to DEI and %WMEN (r = –0.37, P = 0.05; r = –0.44, P = 0.01 respectively; Fig. 2) as well as to percent of calories from fat (r = –0.43, P = 0.02). In both panels of Fig. 2, subjects with negative residual ghrelin values are considered relatively hypoghrelinemic, whereas those with positive residual ghrelin values are considered relatively hyperghrelinemic.
|
|
shows the categorization of subjects according to body weight tertile. Each tertile contained 10 individuals. Whether subjects were categorized in the lower, middle, or upper tertile, relative hyperghrelinemia was associated with lower energy intake expressed as %WMEN, an effect which is especially evident in the leaner people.
|
Fasting plasma insulin concentrations were significantly (P = 0.03) higher in Pima Indians compared with Caucasians (265 ± 10 vs. 210 ± 10 pmol/liter, respectively) and significantly correlated with body weight in both racial groups (r = 0.61, P = 0.02; r = 0.57, P = 0.03; Pima Indians vs. Caucasians, respectively). Fasting plasma insulin concentrations were inversely correlated with fasting plasma ghrelin concentrations in the entire cohort (r = –0.50, P = 0.005) and in Caucasians (r = –0.51, P = 0.05) but not in Pima Indians (r = –0.29, P = 0.29). Additional adjustment of ghrelin for fasting plasma insulin concentrations had no impact on the correlations between ghrelin and energy or fat intake (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Despite recent advances in the understanding of the molecular biology that controls energy homeostasis, the neuroendocrine control of energy intake remains poorly understood, especially in humans. Hormones that regulate food intake can be broadly separated into those that control body fat stores, e.g. insulin and leptin (24, 25, 26), and those that act as short-term satiety signals to initiate or end a meal, e.g. cholecystokinin (27), gut hormone PYY3–36 (28, 29), and ghrelin (9, 10). Ghrelin is believed to act centrally through the stimulation of inhibitory receptors on neuropeptide Y- and Agouti-related protein-producing neurons (6). How and in what proportion of the prevailing circulating levels ghrelin reaches the arcuate nucleus of the hypothalamus to interact with these subpopulations of neurons remain open questions.
Most studies of ghrelin reported thus far have supported the idea that ghrelin is an orexigenic hormone in humans. The most compelling circumstantial evidence to support this notion includes studies showing that, in the fasted state, ghrelin concentrations are elevated and that they rise preprandially and decline postprandially (9, 10), suggesting that the hormone is a short-term satiety factor. More direct evidence comes from a crossover study in which exogenous ghrelin was administered within the physiologic range; in that study, both hunger ratings and food intake increased in all subjects (8). It is difficult, therefore, to reconcile our observations with these results, and we don’t suggest that our data refute the idea that ghrelin is an orexigenic hormone. However, a recent report from the animal literature seems to support our data. Juvenile rats with diet-induced obesity ate more food and gained more weight, but had lower plasma ghrelin concentrations, than their diet-resistant counterparts (30).
One might speculate that the relationship between ghrelin concentrations and intake that we observed was an effect of insulin; however, our results suggest that is not the case because adjusting ghrelin concentrations for insulin concentrations had no impact on the correlations between ghrelin and energy or fat intake. One might also ask whether the observed results are a function of restrained eating behavior; however, once again, our results suggest that this was not the case. In our study, although ghrelin concentrations were measured in the fasted state 2–3 d before the assessment of ad libitum intake, fasting plasma ghrelin reportedly is a good surrogate for 24-h ghrelin concentrations (9). Perhaps, however, a rise in ghrelin concentrations before a meal provides the drive for food intake rather than the average or fasting level. Finally, the problem might be a methodological one, e.g. involving free vs. total ghrelin; our assay procedure measures the total plasma ghrelin concentration, which may or may not correlate with biological activity.
A more attractive conjecture, although perhaps unusual, is that we may be dealing with hyperghrelinemia due to ghrelin resistance. It is difficult to speculate at which level this resistance takes place, because the central regulation of this multihormonal system (including leptin, insulin, and PYY) is exceedingly complex. Although hyperghrelinemia and ghrelin resistance are features associated with cachexia resulting from anorexia (13) as well as systemic disease (31), we are unaware of any other evidence that centrally induced ghrelin resistance results in elevated circulating levels. Vagotomy is associated with elevated plasma ghrelin concentrations (32), modulated by the parasympathetic nervous system (PNS). It is interesting to note that the highest and lowest reported ghrelin concentrations, respectively, have been found in subjects with Prader-Willi syndrome (33), who are known to have low PNS activity (34), and Pima Indians (14), who are known to have high PNS activity (35). In this study, we were unable to establish whether interindividual differences in ghrelinemia are associated with differences in PNS activity.
Another difficulty in interpreting our data comes from the relationship between ghrelin and mean DEI between the races. Although this relationship works very well within race, it is difficult to explain why Caucasians, whose ghrelin concentrations are twice those of Pima Indians, consumed an identical amount of energy.
Prospective studies in free-living conditions testing the role of hyperghrelinemia on future changes in body weight and/or body composition would help to clarify this issue. We are unaware of any such study reported in the literature thus far. However, we previously reported a trend for low, not high, ghrelin concentrations as predictors of future growth in children (15), although those results did not reach statistical significance once the growth variables were adjusted for their respective baseline counterparts. Future studies to confirm or refute these initial results are eagerly anticipated.
In conclusion, this unexpected finding that a low, and not a high, fasting plasma ghrelin concentration predicts ad libitum food intake suggests that the physiologic role of endogenous ghrelin in the regulation of energy homeostasis remains uncertain and raises the possibility that relative hypoghrelinemia not only is associated with obesity but perhaps antecedes it.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received December 15, 2003.
Accepted March 9, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. B Votruba, H. Kirchner, M. Tschop, A. D Salbe, and J. Krakoff Morning ghrelin concentrations are not affected by short-term overfeeding and do not predict ad libitum food intake in humans Am. J. Clinical Nutrition, March 1, 2009; 89(3): 801 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. E Gluck, C. A Venti, A. D Salbe, and J. Krakoff Nighttime eating: commonly observed and related to weight gain in an inpatient food intake study Am. J. Clinical Nutrition, October 1, 2008; 88(4): 900 - 905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Pannacciulli, A. D Salbe, E. Ortega, C. A Venti, C. Bogardus, and J. Krakoff The 24-h carbohydrate oxidation rate in a human respiratory chamber predicts ad libitum food intake Am. J. Clinical Nutrition, September 1, 2007; 86(3): 625 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Marzullo, A. Caumo, G. Savia, B. Verti, G. E. Walker, S. Maestrini, A. Tagliaferri, A. M. Di Blasio, and A. Liuzzi Predictors of Postabsorptive Ghrelin Secretion after Intake of Different Macronutrients J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 4124 - 4130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Bacha and S. A. Arslanian Ghrelin and Peptide YY in Youth: Are There Race-Related Differences? J. Clin. Endocrinol. Metab., August 1, 2006; 91(8): 3117 - 3122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Langenberg, J. Bergstrom, G. A. Laughlin, and E. Barrett-Connor Ghrelin, Adiponectin, and Leptin Do Not Predict Long-term Changes in Weight and Body Mass Index in Older Adults: Longitudinal Analysis of the Rancho Bernardo Cohort Am. J. Epidemiol., December 15, 2005; 162(12): 1189 - 1197. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Langenberg, J. Bergstrom, G. A. Laughlin, and E. Barrett-Connor Ghrelin and the Metabolic Syndrome in Older Adults J. Clin. Endocrinol. Metab., December 1, 2005; 90(12): 6448 - 6453. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) has a high ghrelin value, whereas hypothetical obese subject B (
) and D (
, adjusted values below the regression line) or relatively hyperghrelinemic (
, adjusted values above the regression line). Whether subjects were categorized in the lowest [65 ± 8 kg (range, 51.7–77.4 kg)], middle [91 ± 5 kg (range, 83.7–99.5 kg)], or highest [115 ± 16 kg (range, 99.8–139.8 kg)] tertile of body weight, relative hyperghrelinemia was associated with lower energy intake shown here as %WMEN but true as well for DEI and percent of calories from fat.