Effects of a Fixed Meal Pattern on Ghrelin Secretion: Evidence for a Learned Response Independent of Nutrient Status
http://www.100md.com
《内分泌学杂志》
Department of Psychiatry (D.L.D., R.J.S., S.C.W.), Genome Research Institute
Department of Internal Medicine (T.P.V., D.A.D.), University of Cincinnati, Cincinnati, Ohio 45267
Abstract
Circulating levels of the orexigenic peptide ghrelin increase during fasting and decrease with refeeding. Exogenous ghrelin administration is a potent stimulus for food intake in rodents and humans. In subjects on fixed feeding schedules, ghrelin increases before each meal, raising the possibility that anticipation of meals, in addition to effects of fasting and feeding, contributes to ghrelin secretion. To distinguish among these regulatory influences, plasma ghrelin profiles were generated in freely fed rats and in meal-fed rats trained to consume their daily calories over a 4-h period in the light phase. In freely feeding rats, plasma ghrelin levels increased to a peak of 778 ± 95 pg/ml just before the onset of the dark. Similarly, in meal-fed rats anticipating a large meal of either chow or Ensure at their usual feeding time, plasma ghrelin increased steadily over the 2 h preceding the meal to peaks of 2192 ± 218 and 2075 ± 92 pg/ml, respectively. When freely fed rats were food deprived for a time equivalent to meal-fed rats, there was no peak of plasma ghrelin. In addition, eating-induced suppression of the ghrelin response differed significantly between meal-fed rats and ad libitum-fed rats receiving meals of similar size. These findings indicate that anticipation of eating, as well as fasting/feeding status, influences pre- and postprandial plasma ghrelin levels in rats. Together, these data are consistent with a role for ghrelin in the regulation of anticipatory processes involved in food intake and nutrient disposition.
Introduction
GHRELIN, THE ENDOGENOUS ligand for the GH secretagogue receptor (1), is a 28-amino-acid peptide that is secreted primarily by the stomach and small intestine (2). Ghrelin is a circulating orexigen (reviewed in Ref.3), and food intake increases after administration of exogenous ghrelin in both rodents and humans (4, 5, 6, 7). Consistent with a physiological role for ghrelin in feeding behavior, the administration of antighrelin antibodies or GH secretagogue receptor antagonists reduces food intake (5, 8). The orexigenic action of ghrelin is potent and appears to be mediated, at least in part, through activation of neuropeptide Y/Agouti-related peptide neurons in the arcuate nucleus of the hypothalamus (5, 9, 10). Ghrelin expression in the stomach and ghrelin levels in plasma are elevated after prolonged fasting, and circulating concentrations decrease in response to feeding or the infusion of nutrients. The degree of postprandial ghrelin suppression is a function of the quantity of calories ingested (4, 5, 11). Ghrelin also stimulates gastric motility and gastric acid secretion (12). These findings have led to a consensus that ghrelin is a natural stimulant of food intake (3).
One proposed function of ghrelin is to stimulate hunger and initiate meals. In support of this, when humans are placed on a fixed regimen of three meals at precisely scheduled times each day, plasma ghrelin rises before, and falls soon after, each meal (13). Likewise, sheep maintained on timed feeding schedules have a rise in ghrelin preceding each anticipated meal (14). However, because the human and ovine subjects in these studies had been trained to expect meals at specific times, it is possible that the preprandial rise in plasma ghrelin is part of an anticipatory response, analogous to cephalic insulin (15, 16, 17), rather than a meal initiation signal. In this model, increases in premeal ghrelin occur because the animal is expecting to eat, and the elevated plasma levels regulate processes that are preparatory for the consumption, absorption, and metabolism of a caloric load; in other words, scheduled feeding entrains ghrelin rather than ghrelin dictating feeding. One recent study that sought to distinguish between these alternatives reported that plasma ghrelin profiles were unrelated to the time of spontaneous meal requests by human volunteers maintained in a room devoid of cues related to time or food (11). These findings suggest that, at least in the setting of this experiment, meal initiation is not related to plasma ghrelin levels.
Although it is clear that ghrelin secretion is controlled in part by nutrient status, it is possible that during the preprandial interval, other factors such as learned anticipation contribute to the regulation of ghrelin secretion as well. We designed the following studies to test the hypothesis that the anticipation of a meal is a major determinant of ghrelin secretion. Plasma ghrelin levels were measured in ad libitum, freely fed rats and in meal-fed rats, conditioned over weeks to consume their daily calories in one 4-h period at the same time each day. A group of ad libitum-fed rats that was food deprived for the same amount of time as meal-fed rats before their usual meal was also evaluated. Using this design, we were able to demonstrate that changes in plasma ghrelin levels were related to the anticipation of meals rather than the length of time without food. We also measured the effect of anticipatory factors on ghrelin secretion in the postprandial state using a similar meal-feeding model.
Materials and Methods
Animals
For all experiments, male Long-Evans rats (Harlan, Indianapolis, IN) weighing between 250 and 300 g at the onset of the experiments were used. Before experimental procedures, animals were housed in individual plastic shoebox cages with ad libitum access to pelleted rat chow (Teklad; Harlan, Madison, WI) and water in a temperature-controlled vivarium (22 ± 2 C) on a 12-h light,12-h dark schedule, with lights on at 0700 h. Research was conducted in Association for Assessment and Accreditation of Laboratory Animal Care-approved facilities conforming to National Institutes of Health and U.S. Department of Agriculture regulations, with the approval of the University of Cincinnati Internal Animal Care and Use Committee.
Experiment 1: meal feeding
Rats were assigned to one of two weight-matched groups: an ad libitum-fed group with 24-h access to chow and a meal-fed group with access to chow plus water from 1200–1600 h daily and water alone for the remainder of the day. Food intake was monitored daily in meal-fed rats. After 14 d, the meal-fed rats consumed a similar amount of chow to what they consumed over 24 h before training during the 4-h access period. The same meal-feeding schedule was maintained throughout the experiment, and body weight was recorded daily. In ad libitum-fed animals, food intake and body weight were recorded on d 1 and 14.
To determine the difference in the amount of food that meal-fed animals are capable of consuming compared with ad libitum-fed rats, on d 14, ad libitum- and meal-fed rats had their food removed at 1600 h (the normal time of food removal for the meal-fed rats), and food was returned to all animals at 1200 h the next day. Food intake was recorded at 15, 30, 60, 120, 180, and 240 min after food presentation.
Preprandial ghrelin assessment
Experiment 2a: plasma ghrelin before anticipated chow.
Male rats were meal-fed as described in experiment 1 for 14 d. Food intake was recorded on the final 3 d to ensure that it had reached a plateau. Animals were then weighed and assigned to one of 11 groups with comparable body weights. On d 15, different cohorts were killed by decapitation every 15 min beginning at 1000 h and ending at 1230 h. Blood was collected and plasma stored for measurement of ghrelin, insulin, and glucose.
Among the ad libitum-fed rats, half had food removed at 1600 h on d 14 (i.e. at the same time as the meal-fed rats) and half remained on ad libitum rations. Cohorts (assigned to groups of comparable body weight) were killed every 30 min on d 15 beginning at 1000 h with blood collected identically to the meal-fed animals.
Experiment 2b: plasma ghrelin before anticipated high-fat Ensure Plus.
To generalize the effects of meal feeding among different diets, a separate cohort of rats was meal-fed the palatable, energy-rich diet Ensure Plus (Ross Products Division, Abbott Laboratories, Columbus, OH) as their sole food source, along with ad libitum access to water. Ensure contains 1.5 kcal/ml with 33% of calories as fat, 55% carbohydrate, and 12% protein. In pilot studies, we determined that when rats are meal-fed with Ensure for 3 h a day they consume comparable calories each day as meal-fed animals that have 4 h of access to chow each day. Animals were divided into weight-matched groups, ad libitum and meal-fed. Ad libitum rats had continuous access to Ensure Plus, and meal-fed rats were given access to Ensure from 1200–1500 h each day. Once the Ensure intake of the meal-fed rats reached a plateau, blood was sampled from animals in both groups hourly from 0900–1300 h. None of the meal-fed animals received Ensure Plus on this test day except for the ones sampled at 1300 h.
Experiment 3: plasma ghrelin before dark onset in freely feeding rats.
Rats maintained on ad libitum chow were assigned to one of six groups matched by body weight. Lights went off at 1900 h. Cohorts were killed and blood was collected every 30 min beginning at 1700 h and continuing to 1930 h.
Postprandial ghrelin assessment
Experiment 4a: postprandial ghrelin after an overnight fast in ad libitum-fed rats.
To assess the time course of plasma ghrelin after refeeding, 40 rats were fed chow ad libitum. After 2 wk maintenance, food was removed at 1600 h on one day and returned at 1200 h the next day. Cohorts were killed and blood was collected every 30 min for 2 h, beginning at 1200 h.
Experiment 4b: postprandial ghrelin in meal-fed and ad libitum-fed rats.
Weight-matched rats were assigned to ad libitum- and meal-fed groups consuming chow as in experiment 1. After 10 d of chow, Ensure Plus was substituted for chow as the sole source of calories for both groups. After 6 d of training on Ensure, daily intake of the meal-fed animals had plateaued. On the next day, food was removed from both groups (ad libitum and meal-fed) at 1600 h. In pilot studies, we determined that rats experienced with Ensure Plus drink more than 5 ml within 5 min of presentation. At 1200 h the next day, 5 ml of Ensure Plus was provided to both groups. Animals were killed and blood was sampled at 1200, 1215, 1230, 1300, and 1400 h.
Plasma analysis
Trunk blood was collected in tubes kept on ice and containing heparin (2000 U/ml)/50 mmol/liter EDTA/aprotinin (500 kallikrein inhibitory units/ml). After immediate centrifugation at 3000 rpm, the plasma was stored at –80 C until assayed. Glucose was measured using a glucose oxidase method and insulin was determined by a previously described RIA (18). Plasma ghrelin levels were measured with a commercially available RIA kit (Phoenix Pharmaceuticals Inc., Mountain View, CA).
Statistical analyses
One-way repeated-measures ANOVA was used to analyze meal-fed food intake and body weight in experiment 1. Between-subjects ANOVAs were used where appropriate to determine differences between meal-fed and ad libitum groups. A two-way repeated-measures ANOVA was used for analysis of food intake after deprivation in experiment 1. Plasma hormone data from studies assessing pre- and postprandial ghrelin at various time points were analyzed using one-way ANOVA (in cases where the only comparisons consisted of time) or two-way ANOVA (type of feeding x time point). All pairwise comparisons of mean differences were conducted using Tukey honestly significant differences post hoc comparisons. Differences between group means were considered statistically significant if P < 0.05.
Results
Experiment 1: meal feeding
Before the initiation of meal feeding, rats assigned to this group consumed 21 ± 0.6 g of chow per day. Food intake decreased on the 1st day of meal feeding, but there was a subsequent steady rise in daily consumption as the animals adapted to the regimen. Food intake on d 1 of meal feeding was 6.5 ± 0.3 g compared with 18.5 ± 0.4 g on d 14 (P < 0.05) (Fig. 1A). Ad libitum-fed animals had stable food intake over the 14 d; d 1 food intake was 24.6 ± 0.4 g compared with 25.5 ± 0.4 g on d 14 (P > 0.05). On both d 1 and 14, food intake was significantly greater in ad libitum- than meal-fed animals (P < 0.05). Baseline body weight did not differ between meal-fed and ad libitum-fed animals (meal-fed, 302 ± 3 g; ad libitum, 307 ± 3 g). Both meal-fed and ad libitum-fed rats significantly increased body weight over the course of 14 d; meal-fed rats weighed 322 ± 4 g (Fig. 1B for meal-fed data) and ad libitum-fed rats weighed 368 ± 2 g on d 14 (P < 0.05 for the change in both cases; ad libitum-fed rats were heavier than meal-fed rats; P < 0.05).
Food consumption after 20 h of deprivation was significantly greater for meal-fed than ad libitum-fed rats at all time points after food return (P < 0.05 in all cases) (Fig. 2). Calculated as grams of intake per kilogram of body weight, the difference was even greater.
Experiment 2a: plasma ghrelin before anticipated chow
Plasma ghrelin increased significantly between 1000 and 1130 h in meal-fed rats, starting at a baseline of 1072 ± 63 pg/ml and reaching a peak of 2192 ± 218 pg/ml (Fig. 3A). Plasma ghrelin was significantly elevated above the 1000-h baseline at all time points through 1200 h in meal-fed rats (P < 0.05 for all). At 1145 h, ghrelin values decreased slightly (P > 0.05). After food consumption (7.8 ± 0.2 g at 30 min), ghrelin decreased toward baseline (1151 ± 65 pg/ml at 1230 h) (Fig. 3A).
Ad libitum-fed rats had significantly lower plasma ghrelin levels at all time points compared with the meal-fed rats (e.g. 1000 h, 577 ± 50 pg/ml; P < 0.05) (Fig 3B). In ad libitum-fed rats that had been fasted since 1600 h the previous day, ghrelin levels were significantly higher than those of nonfasted ad libitum-fed rats at all time points (e.g. 1000 h, 1180 ± 119 pg/ml; P < 0.05). Plasma ghrelin was comparable in 20-h fasted meal-fed and 20-h fasted ad libitum-fed rats at the first sampling (P > 0.05). However, fasted ad libitum-rats maintained relatively stable plasma ghrelin over the 2 h before food was presented, and these values were statistically lower than values in meal-fed rats from 1030–1200 h (P < 0.05) (Fig. 3B). After refeeding (1200–1230 h), there was no reduction of plasma ghrelin in the deprived ad libitum-fed group (P > 0.05) despite eating 6.6 ± 0.4 g of food.
Plasma insulin increased steadily between 1000 and 1200 h in the meal-fed rats (from 283 ± 74 pM to 320 ± 37 pM; P < 0.05). Thirty minutes after the onset of feeding, insulin levels were significantly increased compared with preprandial levels (788 ± 120 pM; P < 0.05). Corresponding plasma glucose values were similar before and after the meal: 126 ± 5 mg/dl at 1000 h, 125 ± 4 at 1200 h, and 123 ± 6 at 1230 h. Plasma insulin was relatively constant in the ad libitum-fed and fasted groups throughout the sampling period (e.g. ad libitum-fed at 1000 h, 694 ± 85 pM; ad libitum-fasted at 1000 h, 199 ± 36 pM), although concentrations were significantly reduced in the fasted rats (P < 0.05 vs. ad libitum-fed group at all times). In ad libitum-fasted rats, insulin increased significantly after eating, at 1230 h to 391 ± 48 pM. Plasma glucose followed a similar pattern to the insulin levels. Concentrations changed little from 1000–1200 h but were significantly lower in the ad libitum fasted animals (e.g. ad libitum-fed at 1000 h, 139 ± 3 mg/dl; ad libitum-fasted at 1000 h, 107 ± 4 mg/dl; P < 0.05 at all time points). In ad libitum-fasted rats, glucose increased significantly after eating, at 1230 h to 127 ± 3 mg/dl.
Experiment 2b: plasma ghrelin before anticipated high-fat Ensure Plus
The profile of food intake in meal-fed rats given the liquid meal paralleled the results for solid chow. Ensure Plus intake increased in the meal-fed rats over the course of 14 d (P < 0.05), with intake rising from 10.6 ± 1.8 ml on d 1 of meal feeding to 42.3 ± 1.6 ml on d 14. Ad libitum-fed animals drank 34.2 ± 5.3 ml on d 1 and 76.9 ± 3.4 ml on d 14. Ad libitum-fed rats increased their Ensure Plus intake over the first 4 d, and it then remained stable for the remainder of the study. Animals meal-fed for 3 h with Ensure ate an average of 64.2 kcal at the end of meal feeding, which is comparable to the 66.6 kcal ingested by rats meal-fed chow in 4 h (experiment 1).
Plasma ghrelin increased significantly between 0900 (946 ± 131 pg/ml) and 1200 h (2075 ± 92 pg/ml) in meal-fed rats. Plasma ghrelin was significantly increased at 1200 h compared with all time points preceding it (P < 0.05) (Fig. 4). Postprandial ghrelin (1300 h) was significantly decreased (468 ± 103 pg/ml) compared with levels at 1200 h (P < 0.05) (Fig. 4). Ad libitum-fed animals had stable ghrelin at all time points sampled (0900 h, 1156 ± 169 pg/ml; 1300 h, 917 ± 166 pg/ml; P > 0.05) (Fig. 4).
Insulin increased significantly in meal-fed rats from 0900–1200 h (187 ± 75 to 505 ± 87 pM; P < 0.05), and postprandial insulin (1300 h) was significantly higher still (1796 ± 109 pM; P < 0.05). The corresponding plasma glucose values were 120 ± 4 mg/dl at 0900 h, 131 ± 4 mg/dl at 1200 h, and 134 ± 4 mg/dl at 1300 h (P > 0.05). Ad libitum-fed animals had stable insulin and glucose at all time points assessed (insulin, 422 ± 106 pM at 0900 h, 428 ± 123 pM at 1200 h, and 361 ± 139 pM at 1300 h; glucose, 135 ± 5 mg/dl at 0900 h, 140 ± 6 mg/dl at 1200 h, and 142 ± 5 mg/dl at 1300 h).
Experiment 3: plasma ghrelin before dark onset in freely feeding rats
Ad libitum-fed animals weighed 332 ± 2 g before blood sampling and ate 25 ± 0.3 g of chow daily. Plasma ghrelin rose significantly between 1700 h (606 ± 48 pg/ml) and 1830 h (778 ± 95 pg/ml; P < 0.05) (Fig. 5). Ghrelin decreased slightly but not significantly at 1900 h (lights out). Plasma ghrelin returned to baseline at 1930 h (574 ± 100 ng/dl) (Fig. 5).
Plasma insulin increased significantly before lights out, with a baseline (1700 h) of 346 ± 62 pM and a peak (1830 h) of 569 ± 104 pM (P < 0.05). Postprandial insulin (1930 h) was 680 ± 79 pM. Plasma glucose did not change significantly in the period before lights off (1700 h, 131 ± 6 mg/dl; 1830 h, 136 ± 5 mg/dl), but postprandial glucose (1930 h) increased to 148 ± 7 mg/dl (P < 0.05).
Experiment 4a: postprandial ghrelin after an overnight fast in ad libitum-fed rats
Ad libitum-fed animals that were fasted from 1600 h the previous day and had food returned at 1200 h had ghrelin levels of 1158 ± 67 at 1200 h and did not have a reduction in plasma ghrelin at 1230 h (P > 0.05), but levels were significantly reduced at 1300 h, 1330 h, and 1400 h (P < 0.05 in all cases). These rats ate 4.9 ± 0.3 g of food by 1230 h, 5.9 ± 0.4 g by 1300 h, 6.2 ± 0.3 g at 1330 h, and 7.2 ± 0.6 g at 1400 h (Fig. 6).
Experiment 4b: postprandial ghrelin in meal-fed and ad libitum-fed rats
The plasma ghrelin level of 1527 ± 79 pg/ml at 1200 h in meal-fed animals was not affected by the consumption of 5 ml Ensure Plus. Animals that had been fasted since 1600 h the previous day had ghrelin levels of 1154 ± 150 pg/ml at 1200 h with no change over the first 30 min after drinking 5 ml Ensure. However, ghrelin levels dropped significantly over the subsequent 30 min (859 ± 62 pg/ml at 1300 h; P < 0.05) (Fig 7). Ghrelin values were greater in meal-fed animals compared with ad libitum, fasted animals at all time points assessed (P < 0.05) except at 1215 h.
Discussion
The key finding from these studies is that a major component of ghrelin secretion, both pre- and postprandially, can be entrained by meal feeding independent of the state of fasting or feeding. When rats are habituated to large meals compressed into a 4-h daily period, plasma ghrelin rises steadily over the 2 h preceding the meal, peaking half an hour before the time food is anticipated. In contrast, rats fasted for an equivalent period as the meal-fed animals, but not anticipating a meal, did not have a preprandial rise in ghrelin. Meal-fed rats also differed in the suppression of ghrelin after feeding compared with ad libitum-fed animals. Thus, rats trained to expect their daily rations of chow or Ensure Plus over a short, fixed interval had quantitative and qualitative differences in plasma ghrelin independent of the duration of antecedent fasting or the conditions of refeeding. These findings indicate that a component of ghrelin secretion can be entrained or learned. This response also seems to be present in ad libitum-fed animals in that they have an increase in plasma ghrelin levels before the end of the light period when they normally eat their largest meal of the day (7, 19, 20, 21, 22). Taken together, the results of these experiments support a role for ghrelin in the adaptation that occurs to enable effective meal feeding.
Previous studies of preprandial ghrelin have not controlled for food deprivation. Having a control group that was food deprived for an equal amount of time as the meal-fed rats was essential for demonstrating the learned ghrelin response, given the well known effects of food deprivation per se on ghrelin secretion. When ad libitum-fed rats were food deprived for the same amount of time as meal-fed rats (20 h), they had similarly elevated plasma ghrelin levels relative to nonfasted controls 2 h before the meal but no premeal surge.
The levels of plasma ghrelin were comparable in 18-h fasted rats (i.e. at 1000 h) whether they were meal-fed or ad libitum fed despite the fact that the meal-fed rats weighed significantly less than the ad libitum-fed rats. These data are therefore consistent with reports that fasting increases plasma ghrelin (4), and they suggest that there was no further increase as a result of the weight loss. Although it might be argued that the anticipatory rise of ghrelin observed in the meal-fed rats is caused by the reduced body weight, we consider this unlikely because ad libitum-fed rats with normal body weight that were sampled before lights out also had an anticipatory increase of plasma ghrelin. There are therefore two factors that determine plasma ghrelin before a meal. The first is an increment caused by, and presumably proportional to, the period of food deprivation (5, 23), and the second is a larger increment that occurs when animals are able to anticipate eating. The difference between the two is even more striking considering that the ad libitum-fed rats most likely consumed the majority of their food during the first 4 h of the dark phase the night before the food was removed from their cages. Therefore, these rats consumed their last large meal more than 30 h before the beginning of the blood sampling. Keeping this in mind, plasma ghrelin levels would have been expected to be higher in the ad libitum-fed rats if the duration of food deprivation but not the anticipation of food were the driving factor for the ghrelin secretion. Similarly, the stress for the ad libitum-fed rats who were not used to an overnight food deprivation should have been greater, and ghrelin secretion was found to be increased during water immersion and restraint stress (24).
Collectively, this also speaks to the role of anticipation in the preprandial ghrelin peak observed in our meal-fed rats. One interpretation of the increased ghrelin response in the meal-fed rats is that the increase in circulating ghrelin contributes to the necessary metabolic adjustments the animals need to make to eat their full day’s ration of calories over a short period (16, 25) and that this anticipatory response becomes entrained over 14 d with a regimented meal pattern. Consistent with this interpretation, meal-fed rats eat significantly more food than ad libitum controls deprived for a comparable if not longer period of time. Another interpretation of these data is that both groups of rats were similarly hungry based on hours of deprivation, but the meal-fed group was able to eat more food than the ad libitum-fasted rats because of other meal-preparatory functions of ghrelin, beyond its orexigenic properties.
In meal-fed rats expecting their daily food, and in ad libitum-fed rats before lights out, plasma ghrelin increases to a peak around 30 min before the anticipated meal and then declines somewhat. This is consistent with previous results in meal-fed sheep (14). If there were a direct correspondence between plasma ghrelin and meal initiation, the ad libitum-fed rats should begin eating before lights out, at the time of the ghrelin peak, and in meal-fed rats, the peak should occur at the time food is expected rather than 30 min before. That they do not raises some question about the role of ghrelin in meal initiation. Furthermore, the fact that meal-fed animals eat steadily over the course of 4 h, consuming approximately 5 g in the first 15 min and approximately 19 g after 4 h despite a steep drop in plasma ghrelin, is also inconsistent with a purely orexigenic function. A more likely explanation is that the rise in circulating ghrelin is tied to the anticipation of feeding at a predictable time of day. In this setting, ghrelin may function to initiate preparatory responses that enable the animal to consume a large caloric load without significant disruption of nutrient homeostasis; previously demonstrated ghrelin effects on gastrointestinal or endocrine pancreatic function are potential sites of this action (26, 27, 28, 29, 30). This role would be analogous to, or possibly synergistic with, other anticipatory responses such as premeal, or cephalic, insulin secretion.
It is noteworthy that the preprandial ghrelin peak was considerably larger in meal-fed rats anticipating their daily 4-h period of food availability compared with ad libitum-fed rats before the onset of the dark. One interpretation of these data is that the higher ghrelin peak observed in the meal-fed rats is necessary to prepare them for the intake of a large amount of calories in a short window of time. Second, ad libitum-fed rats most likely start to eat at varying time points around the onset of the dark phase. Therefore, one could expect that their maximum plasma ghrelin level occurs with a larger temporal variance than occurs in the highly trained meal-fed rats that receive their food precisely at 1200 h every day. Although we have no direct evidence to rule out the possibility that the preprandial peak in plasma ghrelin in meal-fed rats was simply a strong orexigenic stimulus that elicits the increased food intake during their 4 h of food access, we do not consider this to be likely. Rather, we believe that the increment of ghrelin reflects the fact that more food will be consumed, and that it assists in preparing the body to receive the calories; however, additional studies, perhaps using GH secretagogue antagonists, will be necessary to gain additional insight into these questions.
Because eating large meals is a necessity of life in some environments, an important strategy for minimizing their impact is to initiate anticipatory responses before the meal that lessen the postprandial perturbation (reviewed in Refs.16 and 25). The best studied of these anticipatory responses is the premeal secretion of insulin to minimize the prandial increases in blood glucose (17). The phenomenon of cephalic insulin has been demonstrated in both rats (31) and humans (15, 32), and if cephalic insulin secretion is prevented or its action blocked, smaller meals are consumed (33). There are numerous other meal-anticipatory responses including a premeal reduction in metabolic rate, a premeal elevation of plasma corticosterone, and a premeal increase in body temperature (34, 35, 36). Interestingly, the preprandial rise of corticosterone is comparable to the increase in ghrelin. The involvement of both hormones in the regulation of food intake and energy balance warrants the exploration of the precise temporal relationship between these two hormones in future studies as well as how they may interact to prepare an individual for an imminent meal.
In addition to its effects on preprandial ghrelin, meal anticipation also had profound effects on postprandial ghrelin. In all previous studies to date, when food-deprived individuals ingest nutrients, plasma ghrelin drops rapidly (11, 13, 14); our data from experiment 4a in ad libitum-fed rats are consistent with this. However, when meal-fed rats and ad libitum-fed rats that were similarly food deprived subsequently ingested the same fixed amount of food, the postprandial ghrelin responses of these groups were dramatically different. Only the ad libitum deprived rats showed reduced ghrelin in response to their 5-ml meal of Ensure Plus, with plasma ghrelin returning to prefasting baseline values within 1 h. Meal-fed rats, on the other hand, after ingesting the same 5-ml meal showed no change of plasma ghrelin over the course of the 2 h assessed. The present findings suggest that anticipating a particular number of calories can influence the postprandial ghrelin response. In this study, meal-fed rats consumed only 12% of what they had been trained to expect over that time period and therefore did not suppress ghrelin; the identical number of calories completely suppressed ghrelin in ad libitum-fed rats deprived the same amount of time as meal-fed rats. Previous studies have shown that postprandial ghrelin suppression is directly proportional to calories ingested (11, 37). Our data are the first to demonstrate that postprandial ghrelin, similar to preprandial ghrelin, is dependent on meal anticipation as well as nutrient status.
Our data in rats appear to conflict with a report in humans suggesting that ghrelin functions as an appetite signal. In that study, plasma ghrelin levels rose before voluntary meal requests in subjects devoid of time or food cues, and this preprandial ghrelin rise was correlated with subjective hunger scores (38). However, it should be noted that the subjects in that study were accustomed to eating three meals per day, and given that they had been deprived of time cues for only a few hours, it is possible that their endogenous patterns of feeding behavior persisted. Thus, it is likely that anticipatory influences on ghrelin secretion are also present in humans, although this question will require rigorous examination.
In summary, the present results demonstrate that ghrelin secretion is regulated by learned anticipation independent of deprivation, in addition to regulation by nutrient status. Consistent with previous data from other species including humans, there is a learned, preprandial increase of plasma ghrelin when individuals have been trained to anticipate their food at a fixed time. When food deprivation was held constant, fasted ad libitum-fed rats maintained high, consistent ghrelin values over time, whereas meal-fed rats displayed a rise and a prominent peak of plasma ghrelin 30 min before food was expected. These data, when considered in the context of previous data, support a potential role for ghrelin in anticipatory meal preparation. Preprandial ghrelin may serve to prepare the stomach and gastrointestinal tract for anticipated food. Anticipation of food also had an impact on postprandial ghrelin, suggesting that the postprandial ghrelin response is not simply related to the number of calories ingested. Future studies are necessary to determine the mechanisms by which anticipation interacts with nutrient status to affect both pre- and postprandial ghrelin.
Acknowledgments
We thank Kay Ellis for her careful assay of plasma hormones and Kathi Smith, Joyce Sorrell, Eileen Elfers, Ken Parks, and Mouhamadoul Toure for technical assistance.
Footnotes
This work was supported by several grants from the National Institutes of Health: DK 63779 (D.L.D.), 17844 (S.C.W.), 56863 (D.A.D.), and 54080 (R.J.S.). The Obesity Research Center at the University of Cincinnati is supported in part by Procter & Gamble.
First Published Online September 22, 2005
Accepted for publication September 8, 2005.
References
Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660
Date Y, Kojima M, Hosoda H, Sawaguchi A, Mondal MS, Suganuma T, Matsukura S, Kangawa K, Nakazato M 2000 Ghrelin, a novel growth hormone-releasing acylated peptide, is synthesized in a distinct endocrine cell type in the gastrointestinal tracts of rats and humans. Endocrinology 141:4255–4261
Cummings DE, Shannon MH 2003 Roles for ghrelin in the regulation of appetite and body weight. Arch Surg 138:389–396
Tschop M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913
Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S 2001 A role for ghrelin in the central regulation of feeding. Nature 409:194–198
Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR 2001 Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992
Tolle V, Bassant MH, Zizzari P, Poindessous-Jazat F, Tomasetto C, Epelbaum J, Bluet-Pajot MT 2002 Ultradian rhythmicity of ghrelin secretion in relation with GH, feeding behavior, and sleep-wake patterns in rats. Endocrinology 143:1353–1361
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Fujimiya M, Katsuura G, Makino S, Fujino MA, Kasuga M 2001 A role of ghrelin in neuroendocrine and behavioral responses to stress in mice. Neuroendocrinology 74:143–147
Shintani M, Ogawa Y, Ebihara K, Aizawa-Abe M, Miyanaga F, Takaya K, Hayashi T, Inoue G, Hosoda K, Kojima M, Kangawa K, Nakao K 2001 Ghrelin, an endogenous growth hormone secretagogue, is a novel orexigenic peptide that antagonizes leptin action through the activation of hypothalamic neuropeptide Y/Y1 receptor pathway. Diabetes 50:227–232
Kamegai J, Tamura H, Shimizu T, Ishii S, Sugihara H, Wakabayashi I 2000 Central effect of ghrelin, an endogenous growth hormone secretagogue, on hypothalamic peptide gene expression. Endocrinology 141:4797–4800
Callahan HS, Cummings DE, Pepe MS, Breen PA, Matthys CC, Weigle DS 2004 Postprandial suppression of plasma ghrelin level is proportional to ingested caloric load but does not predict intermeal interval in humans. J Clin Endocrinol Metab 89:1319–1324
Masuda Y, Tanaka T, Inomata N, Ohnuma N, Tanaka S, Itoh Z, Hosoda H, Kojima M, Kangawa K 2000 Ghrelin stimulates gastric acid secretion and motility in rats. Biochem Biophys Res Commun 276:905–908
Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse BE, Weigle DS 2001 A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 50:1714–1719
Sugino T, Hasegawa Y, Kikkawa Y, Yamaura J, Yamagishi M, Kurose Y, Kojima M, Kangawa K, Terashima Y 2002 A transient ghrelin surge occurs just before feeding in a scheduled meal-fed sheep. Biochem Biophys Res Commun 295:255–260
Teff K 2000 Nutritional implications of the cephalic-phase reflexes: endocrine responses. Appetite 34:206–213
Woods SC 1991 The eating paradox: how we tolerate food. Psychol Rev 98:488–505
Berthoud H-R, Berieter DA, Trimble ER, Siegel EG, Jeanrenaud B 1981 Cephalic phase, reflex insulin secretion: neuroanatomical and physiological characterization. Diabetologia 20:393–401
Ensinck JW, Laschansky EC, Vogel RE, D’Alessio DA 1991 Effect of somatostatin-28 on dynamics of insulin secretion in perfused rat pancreas. Diabetes 40:1163–1169
Kissileff HR 1970 Free feeding in normal and "recovered lateral" rats monitored by a pellet-detecting eatometer. Physiol Behav 5:163–173
Strubbe JH, Keyser J, Dijkstra T, Prins AJ 1986 Interaction between circadian and caloric control of feeding behavior in the rat. Physiol Behav 36:489–493
Murakami N, Hayashida T, Kuroiwa T, Nakahara K, Ida T, Mondal MS, Nakazato M, Kojima M, Kangawa K 2002 Role for central ghrelin in food intake and secretion profile of stomach ghrelin in rats. J Endocrinol 174:283–288
Strubbe JH, Woods SC 2004 The timing of meals. Psychol Rev 111:128–141
Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger EP, Purnell JQ 2002 Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 346:1623–1630
Brzozowski T, Konturek PC, Konturek SJ, Kwiecien S, Drozdowicz D, Bielanski W, Pajdo R, Ptak A, Nikiforuk A, Pawlik WW, Hahn EG 2004 Exogenous and endogenous ghrelin in gastroprotection against stress-induced gastric damage. Regul Pept 120:39–51
Woods SC, Strubbe JH 1994 The psychobiology of meals. Psychonomic Bulletin and Review 1:141–155
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Ueno N, Makino S, Fujimiya M, Niijima A, Fujino MA, Kasuga M 2001 Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 120:337–345
Trudel L, Tomasetto C, Rio MC, Bouin M, Plourde V, Eberling P, Poitras P 2002 Ghrelin/motilin-related peptide is a potent prokinetic to reverse gastric postoperative ileus in rat. Am J Physiol Gastrointest Liver Physiol 282:G948–G952
Inui A, Asakawa A, Bowers CY, Mantovani G, Laviano A, Meguid MM, Fujimiya M 2004 Ghrelin, appetite, and gastric motility: the emerging role of the stomach as an endocrine organ. FASEB J 18:439–456
Lee HM, Wang G, Englander EW, Kojima M, Greeley Jr GH 2002 Ghrelin, a new gastrointestinal endocrine peptide that stimulates insulin secretion: enteric distribution, ontogeny, influence of endocrine, and dietary manipulations. Endocrinology 143:185–190
Reimer MK, Pacini G, Ahren B 2003 Dose-dependent inhibition by ghrelin of insulin secretion in the mouse. Endocrinology 144:916–921
Strubbe JH, Steffens AB 1975 Rapid insulin release after ingestion of a meal in the unanesthetized rat. Am J Physiol 229:1019–1022
Teff KL, Engelman K 1996 Palatability and dietary restraint: effect on cephalic phase insulin release in women. Physiol Behav 60:567–573
Louis-Sylvestre J 1976 Preabsorptive insulin release and hypoglycemia in rats. Am J Physiol 230:56–60
Rowe JW, Young JB, Minaker KL, Stevens AL, Pallotta J, Landsberg L 1981 Effect of insulin and glucose infusions on sympathetic nervous system activity in normal man. Diabetes 30:219–225
Powley TL, Berthoud HR 1985 Diet and cephalic phase insulin responses. Am J Clin Nutr 42:991–1002
de Vries J, Strubbe JH, Wildering WC, Gorter JA, Prins AJA 1993 Patterns of body temperature during feeding in rats under varying ambient temperatures. Physiol Behav 53:229–235
le Roux CW, Patterson M, Vincent RP, Hunt C, Ghatei MA, Bloom SR 2005 Postprandial plasma ghrelin is suppressed proportional to meal calorie content in normal-weight but not obese subjects. J Clin Endocrinol Metab 90:1068–1071
Cummings DE, Frayo RS, Marmonier C, Aubert R, Chapelot D 2004 Plasma ghrelin levels and hunger scores in humans initiating meals voluntarily without time- and food-related cues. The adipoinsular axis: effects of leptin on pancreatic -cells 287:E297–E304(Deborah L. Drazen, Torsten P. Vahl, Davi)
Department of Internal Medicine (T.P.V., D.A.D.), University of Cincinnati, Cincinnati, Ohio 45267
Abstract
Circulating levels of the orexigenic peptide ghrelin increase during fasting and decrease with refeeding. Exogenous ghrelin administration is a potent stimulus for food intake in rodents and humans. In subjects on fixed feeding schedules, ghrelin increases before each meal, raising the possibility that anticipation of meals, in addition to effects of fasting and feeding, contributes to ghrelin secretion. To distinguish among these regulatory influences, plasma ghrelin profiles were generated in freely fed rats and in meal-fed rats trained to consume their daily calories over a 4-h period in the light phase. In freely feeding rats, plasma ghrelin levels increased to a peak of 778 ± 95 pg/ml just before the onset of the dark. Similarly, in meal-fed rats anticipating a large meal of either chow or Ensure at their usual feeding time, plasma ghrelin increased steadily over the 2 h preceding the meal to peaks of 2192 ± 218 and 2075 ± 92 pg/ml, respectively. When freely fed rats were food deprived for a time equivalent to meal-fed rats, there was no peak of plasma ghrelin. In addition, eating-induced suppression of the ghrelin response differed significantly between meal-fed rats and ad libitum-fed rats receiving meals of similar size. These findings indicate that anticipation of eating, as well as fasting/feeding status, influences pre- and postprandial plasma ghrelin levels in rats. Together, these data are consistent with a role for ghrelin in the regulation of anticipatory processes involved in food intake and nutrient disposition.
Introduction
GHRELIN, THE ENDOGENOUS ligand for the GH secretagogue receptor (1), is a 28-amino-acid peptide that is secreted primarily by the stomach and small intestine (2). Ghrelin is a circulating orexigen (reviewed in Ref.3), and food intake increases after administration of exogenous ghrelin in both rodents and humans (4, 5, 6, 7). Consistent with a physiological role for ghrelin in feeding behavior, the administration of antighrelin antibodies or GH secretagogue receptor antagonists reduces food intake (5, 8). The orexigenic action of ghrelin is potent and appears to be mediated, at least in part, through activation of neuropeptide Y/Agouti-related peptide neurons in the arcuate nucleus of the hypothalamus (5, 9, 10). Ghrelin expression in the stomach and ghrelin levels in plasma are elevated after prolonged fasting, and circulating concentrations decrease in response to feeding or the infusion of nutrients. The degree of postprandial ghrelin suppression is a function of the quantity of calories ingested (4, 5, 11). Ghrelin also stimulates gastric motility and gastric acid secretion (12). These findings have led to a consensus that ghrelin is a natural stimulant of food intake (3).
One proposed function of ghrelin is to stimulate hunger and initiate meals. In support of this, when humans are placed on a fixed regimen of three meals at precisely scheduled times each day, plasma ghrelin rises before, and falls soon after, each meal (13). Likewise, sheep maintained on timed feeding schedules have a rise in ghrelin preceding each anticipated meal (14). However, because the human and ovine subjects in these studies had been trained to expect meals at specific times, it is possible that the preprandial rise in plasma ghrelin is part of an anticipatory response, analogous to cephalic insulin (15, 16, 17), rather than a meal initiation signal. In this model, increases in premeal ghrelin occur because the animal is expecting to eat, and the elevated plasma levels regulate processes that are preparatory for the consumption, absorption, and metabolism of a caloric load; in other words, scheduled feeding entrains ghrelin rather than ghrelin dictating feeding. One recent study that sought to distinguish between these alternatives reported that plasma ghrelin profiles were unrelated to the time of spontaneous meal requests by human volunteers maintained in a room devoid of cues related to time or food (11). These findings suggest that, at least in the setting of this experiment, meal initiation is not related to plasma ghrelin levels.
Although it is clear that ghrelin secretion is controlled in part by nutrient status, it is possible that during the preprandial interval, other factors such as learned anticipation contribute to the regulation of ghrelin secretion as well. We designed the following studies to test the hypothesis that the anticipation of a meal is a major determinant of ghrelin secretion. Plasma ghrelin levels were measured in ad libitum, freely fed rats and in meal-fed rats, conditioned over weeks to consume their daily calories in one 4-h period at the same time each day. A group of ad libitum-fed rats that was food deprived for the same amount of time as meal-fed rats before their usual meal was also evaluated. Using this design, we were able to demonstrate that changes in plasma ghrelin levels were related to the anticipation of meals rather than the length of time without food. We also measured the effect of anticipatory factors on ghrelin secretion in the postprandial state using a similar meal-feeding model.
Materials and Methods
Animals
For all experiments, male Long-Evans rats (Harlan, Indianapolis, IN) weighing between 250 and 300 g at the onset of the experiments were used. Before experimental procedures, animals were housed in individual plastic shoebox cages with ad libitum access to pelleted rat chow (Teklad; Harlan, Madison, WI) and water in a temperature-controlled vivarium (22 ± 2 C) on a 12-h light,12-h dark schedule, with lights on at 0700 h. Research was conducted in Association for Assessment and Accreditation of Laboratory Animal Care-approved facilities conforming to National Institutes of Health and U.S. Department of Agriculture regulations, with the approval of the University of Cincinnati Internal Animal Care and Use Committee.
Experiment 1: meal feeding
Rats were assigned to one of two weight-matched groups: an ad libitum-fed group with 24-h access to chow and a meal-fed group with access to chow plus water from 1200–1600 h daily and water alone for the remainder of the day. Food intake was monitored daily in meal-fed rats. After 14 d, the meal-fed rats consumed a similar amount of chow to what they consumed over 24 h before training during the 4-h access period. The same meal-feeding schedule was maintained throughout the experiment, and body weight was recorded daily. In ad libitum-fed animals, food intake and body weight were recorded on d 1 and 14.
To determine the difference in the amount of food that meal-fed animals are capable of consuming compared with ad libitum-fed rats, on d 14, ad libitum- and meal-fed rats had their food removed at 1600 h (the normal time of food removal for the meal-fed rats), and food was returned to all animals at 1200 h the next day. Food intake was recorded at 15, 30, 60, 120, 180, and 240 min after food presentation.
Preprandial ghrelin assessment
Experiment 2a: plasma ghrelin before anticipated chow.
Male rats were meal-fed as described in experiment 1 for 14 d. Food intake was recorded on the final 3 d to ensure that it had reached a plateau. Animals were then weighed and assigned to one of 11 groups with comparable body weights. On d 15, different cohorts were killed by decapitation every 15 min beginning at 1000 h and ending at 1230 h. Blood was collected and plasma stored for measurement of ghrelin, insulin, and glucose.
Among the ad libitum-fed rats, half had food removed at 1600 h on d 14 (i.e. at the same time as the meal-fed rats) and half remained on ad libitum rations. Cohorts (assigned to groups of comparable body weight) were killed every 30 min on d 15 beginning at 1000 h with blood collected identically to the meal-fed animals.
Experiment 2b: plasma ghrelin before anticipated high-fat Ensure Plus.
To generalize the effects of meal feeding among different diets, a separate cohort of rats was meal-fed the palatable, energy-rich diet Ensure Plus (Ross Products Division, Abbott Laboratories, Columbus, OH) as their sole food source, along with ad libitum access to water. Ensure contains 1.5 kcal/ml with 33% of calories as fat, 55% carbohydrate, and 12% protein. In pilot studies, we determined that when rats are meal-fed with Ensure for 3 h a day they consume comparable calories each day as meal-fed animals that have 4 h of access to chow each day. Animals were divided into weight-matched groups, ad libitum and meal-fed. Ad libitum rats had continuous access to Ensure Plus, and meal-fed rats were given access to Ensure from 1200–1500 h each day. Once the Ensure intake of the meal-fed rats reached a plateau, blood was sampled from animals in both groups hourly from 0900–1300 h. None of the meal-fed animals received Ensure Plus on this test day except for the ones sampled at 1300 h.
Experiment 3: plasma ghrelin before dark onset in freely feeding rats.
Rats maintained on ad libitum chow were assigned to one of six groups matched by body weight. Lights went off at 1900 h. Cohorts were killed and blood was collected every 30 min beginning at 1700 h and continuing to 1930 h.
Postprandial ghrelin assessment
Experiment 4a: postprandial ghrelin after an overnight fast in ad libitum-fed rats.
To assess the time course of plasma ghrelin after refeeding, 40 rats were fed chow ad libitum. After 2 wk maintenance, food was removed at 1600 h on one day and returned at 1200 h the next day. Cohorts were killed and blood was collected every 30 min for 2 h, beginning at 1200 h.
Experiment 4b: postprandial ghrelin in meal-fed and ad libitum-fed rats.
Weight-matched rats were assigned to ad libitum- and meal-fed groups consuming chow as in experiment 1. After 10 d of chow, Ensure Plus was substituted for chow as the sole source of calories for both groups. After 6 d of training on Ensure, daily intake of the meal-fed animals had plateaued. On the next day, food was removed from both groups (ad libitum and meal-fed) at 1600 h. In pilot studies, we determined that rats experienced with Ensure Plus drink more than 5 ml within 5 min of presentation. At 1200 h the next day, 5 ml of Ensure Plus was provided to both groups. Animals were killed and blood was sampled at 1200, 1215, 1230, 1300, and 1400 h.
Plasma analysis
Trunk blood was collected in tubes kept on ice and containing heparin (2000 U/ml)/50 mmol/liter EDTA/aprotinin (500 kallikrein inhibitory units/ml). After immediate centrifugation at 3000 rpm, the plasma was stored at –80 C until assayed. Glucose was measured using a glucose oxidase method and insulin was determined by a previously described RIA (18). Plasma ghrelin levels were measured with a commercially available RIA kit (Phoenix Pharmaceuticals Inc., Mountain View, CA).
Statistical analyses
One-way repeated-measures ANOVA was used to analyze meal-fed food intake and body weight in experiment 1. Between-subjects ANOVAs were used where appropriate to determine differences between meal-fed and ad libitum groups. A two-way repeated-measures ANOVA was used for analysis of food intake after deprivation in experiment 1. Plasma hormone data from studies assessing pre- and postprandial ghrelin at various time points were analyzed using one-way ANOVA (in cases where the only comparisons consisted of time) or two-way ANOVA (type of feeding x time point). All pairwise comparisons of mean differences were conducted using Tukey honestly significant differences post hoc comparisons. Differences between group means were considered statistically significant if P < 0.05.
Results
Experiment 1: meal feeding
Before the initiation of meal feeding, rats assigned to this group consumed 21 ± 0.6 g of chow per day. Food intake decreased on the 1st day of meal feeding, but there was a subsequent steady rise in daily consumption as the animals adapted to the regimen. Food intake on d 1 of meal feeding was 6.5 ± 0.3 g compared with 18.5 ± 0.4 g on d 14 (P < 0.05) (Fig. 1A). Ad libitum-fed animals had stable food intake over the 14 d; d 1 food intake was 24.6 ± 0.4 g compared with 25.5 ± 0.4 g on d 14 (P > 0.05). On both d 1 and 14, food intake was significantly greater in ad libitum- than meal-fed animals (P < 0.05). Baseline body weight did not differ between meal-fed and ad libitum-fed animals (meal-fed, 302 ± 3 g; ad libitum, 307 ± 3 g). Both meal-fed and ad libitum-fed rats significantly increased body weight over the course of 14 d; meal-fed rats weighed 322 ± 4 g (Fig. 1B for meal-fed data) and ad libitum-fed rats weighed 368 ± 2 g on d 14 (P < 0.05 for the change in both cases; ad libitum-fed rats were heavier than meal-fed rats; P < 0.05).
Food consumption after 20 h of deprivation was significantly greater for meal-fed than ad libitum-fed rats at all time points after food return (P < 0.05 in all cases) (Fig. 2). Calculated as grams of intake per kilogram of body weight, the difference was even greater.
Experiment 2a: plasma ghrelin before anticipated chow
Plasma ghrelin increased significantly between 1000 and 1130 h in meal-fed rats, starting at a baseline of 1072 ± 63 pg/ml and reaching a peak of 2192 ± 218 pg/ml (Fig. 3A). Plasma ghrelin was significantly elevated above the 1000-h baseline at all time points through 1200 h in meal-fed rats (P < 0.05 for all). At 1145 h, ghrelin values decreased slightly (P > 0.05). After food consumption (7.8 ± 0.2 g at 30 min), ghrelin decreased toward baseline (1151 ± 65 pg/ml at 1230 h) (Fig. 3A).
Ad libitum-fed rats had significantly lower plasma ghrelin levels at all time points compared with the meal-fed rats (e.g. 1000 h, 577 ± 50 pg/ml; P < 0.05) (Fig 3B). In ad libitum-fed rats that had been fasted since 1600 h the previous day, ghrelin levels were significantly higher than those of nonfasted ad libitum-fed rats at all time points (e.g. 1000 h, 1180 ± 119 pg/ml; P < 0.05). Plasma ghrelin was comparable in 20-h fasted meal-fed and 20-h fasted ad libitum-fed rats at the first sampling (P > 0.05). However, fasted ad libitum-rats maintained relatively stable plasma ghrelin over the 2 h before food was presented, and these values were statistically lower than values in meal-fed rats from 1030–1200 h (P < 0.05) (Fig. 3B). After refeeding (1200–1230 h), there was no reduction of plasma ghrelin in the deprived ad libitum-fed group (P > 0.05) despite eating 6.6 ± 0.4 g of food.
Plasma insulin increased steadily between 1000 and 1200 h in the meal-fed rats (from 283 ± 74 pM to 320 ± 37 pM; P < 0.05). Thirty minutes after the onset of feeding, insulin levels were significantly increased compared with preprandial levels (788 ± 120 pM; P < 0.05). Corresponding plasma glucose values were similar before and after the meal: 126 ± 5 mg/dl at 1000 h, 125 ± 4 at 1200 h, and 123 ± 6 at 1230 h. Plasma insulin was relatively constant in the ad libitum-fed and fasted groups throughout the sampling period (e.g. ad libitum-fed at 1000 h, 694 ± 85 pM; ad libitum-fasted at 1000 h, 199 ± 36 pM), although concentrations were significantly reduced in the fasted rats (P < 0.05 vs. ad libitum-fed group at all times). In ad libitum-fasted rats, insulin increased significantly after eating, at 1230 h to 391 ± 48 pM. Plasma glucose followed a similar pattern to the insulin levels. Concentrations changed little from 1000–1200 h but were significantly lower in the ad libitum fasted animals (e.g. ad libitum-fed at 1000 h, 139 ± 3 mg/dl; ad libitum-fasted at 1000 h, 107 ± 4 mg/dl; P < 0.05 at all time points). In ad libitum-fasted rats, glucose increased significantly after eating, at 1230 h to 127 ± 3 mg/dl.
Experiment 2b: plasma ghrelin before anticipated high-fat Ensure Plus
The profile of food intake in meal-fed rats given the liquid meal paralleled the results for solid chow. Ensure Plus intake increased in the meal-fed rats over the course of 14 d (P < 0.05), with intake rising from 10.6 ± 1.8 ml on d 1 of meal feeding to 42.3 ± 1.6 ml on d 14. Ad libitum-fed animals drank 34.2 ± 5.3 ml on d 1 and 76.9 ± 3.4 ml on d 14. Ad libitum-fed rats increased their Ensure Plus intake over the first 4 d, and it then remained stable for the remainder of the study. Animals meal-fed for 3 h with Ensure ate an average of 64.2 kcal at the end of meal feeding, which is comparable to the 66.6 kcal ingested by rats meal-fed chow in 4 h (experiment 1).
Plasma ghrelin increased significantly between 0900 (946 ± 131 pg/ml) and 1200 h (2075 ± 92 pg/ml) in meal-fed rats. Plasma ghrelin was significantly increased at 1200 h compared with all time points preceding it (P < 0.05) (Fig. 4). Postprandial ghrelin (1300 h) was significantly decreased (468 ± 103 pg/ml) compared with levels at 1200 h (P < 0.05) (Fig. 4). Ad libitum-fed animals had stable ghrelin at all time points sampled (0900 h, 1156 ± 169 pg/ml; 1300 h, 917 ± 166 pg/ml; P > 0.05) (Fig. 4).
Insulin increased significantly in meal-fed rats from 0900–1200 h (187 ± 75 to 505 ± 87 pM; P < 0.05), and postprandial insulin (1300 h) was significantly higher still (1796 ± 109 pM; P < 0.05). The corresponding plasma glucose values were 120 ± 4 mg/dl at 0900 h, 131 ± 4 mg/dl at 1200 h, and 134 ± 4 mg/dl at 1300 h (P > 0.05). Ad libitum-fed animals had stable insulin and glucose at all time points assessed (insulin, 422 ± 106 pM at 0900 h, 428 ± 123 pM at 1200 h, and 361 ± 139 pM at 1300 h; glucose, 135 ± 5 mg/dl at 0900 h, 140 ± 6 mg/dl at 1200 h, and 142 ± 5 mg/dl at 1300 h).
Experiment 3: plasma ghrelin before dark onset in freely feeding rats
Ad libitum-fed animals weighed 332 ± 2 g before blood sampling and ate 25 ± 0.3 g of chow daily. Plasma ghrelin rose significantly between 1700 h (606 ± 48 pg/ml) and 1830 h (778 ± 95 pg/ml; P < 0.05) (Fig. 5). Ghrelin decreased slightly but not significantly at 1900 h (lights out). Plasma ghrelin returned to baseline at 1930 h (574 ± 100 ng/dl) (Fig. 5).
Plasma insulin increased significantly before lights out, with a baseline (1700 h) of 346 ± 62 pM and a peak (1830 h) of 569 ± 104 pM (P < 0.05). Postprandial insulin (1930 h) was 680 ± 79 pM. Plasma glucose did not change significantly in the period before lights off (1700 h, 131 ± 6 mg/dl; 1830 h, 136 ± 5 mg/dl), but postprandial glucose (1930 h) increased to 148 ± 7 mg/dl (P < 0.05).
Experiment 4a: postprandial ghrelin after an overnight fast in ad libitum-fed rats
Ad libitum-fed animals that were fasted from 1600 h the previous day and had food returned at 1200 h had ghrelin levels of 1158 ± 67 at 1200 h and did not have a reduction in plasma ghrelin at 1230 h (P > 0.05), but levels were significantly reduced at 1300 h, 1330 h, and 1400 h (P < 0.05 in all cases). These rats ate 4.9 ± 0.3 g of food by 1230 h, 5.9 ± 0.4 g by 1300 h, 6.2 ± 0.3 g at 1330 h, and 7.2 ± 0.6 g at 1400 h (Fig. 6).
Experiment 4b: postprandial ghrelin in meal-fed and ad libitum-fed rats
The plasma ghrelin level of 1527 ± 79 pg/ml at 1200 h in meal-fed animals was not affected by the consumption of 5 ml Ensure Plus. Animals that had been fasted since 1600 h the previous day had ghrelin levels of 1154 ± 150 pg/ml at 1200 h with no change over the first 30 min after drinking 5 ml Ensure. However, ghrelin levels dropped significantly over the subsequent 30 min (859 ± 62 pg/ml at 1300 h; P < 0.05) (Fig 7). Ghrelin values were greater in meal-fed animals compared with ad libitum, fasted animals at all time points assessed (P < 0.05) except at 1215 h.
Discussion
The key finding from these studies is that a major component of ghrelin secretion, both pre- and postprandially, can be entrained by meal feeding independent of the state of fasting or feeding. When rats are habituated to large meals compressed into a 4-h daily period, plasma ghrelin rises steadily over the 2 h preceding the meal, peaking half an hour before the time food is anticipated. In contrast, rats fasted for an equivalent period as the meal-fed animals, but not anticipating a meal, did not have a preprandial rise in ghrelin. Meal-fed rats also differed in the suppression of ghrelin after feeding compared with ad libitum-fed animals. Thus, rats trained to expect their daily rations of chow or Ensure Plus over a short, fixed interval had quantitative and qualitative differences in plasma ghrelin independent of the duration of antecedent fasting or the conditions of refeeding. These findings indicate that a component of ghrelin secretion can be entrained or learned. This response also seems to be present in ad libitum-fed animals in that they have an increase in plasma ghrelin levels before the end of the light period when they normally eat their largest meal of the day (7, 19, 20, 21, 22). Taken together, the results of these experiments support a role for ghrelin in the adaptation that occurs to enable effective meal feeding.
Previous studies of preprandial ghrelin have not controlled for food deprivation. Having a control group that was food deprived for an equal amount of time as the meal-fed rats was essential for demonstrating the learned ghrelin response, given the well known effects of food deprivation per se on ghrelin secretion. When ad libitum-fed rats were food deprived for the same amount of time as meal-fed rats (20 h), they had similarly elevated plasma ghrelin levels relative to nonfasted controls 2 h before the meal but no premeal surge.
The levels of plasma ghrelin were comparable in 18-h fasted rats (i.e. at 1000 h) whether they were meal-fed or ad libitum fed despite the fact that the meal-fed rats weighed significantly less than the ad libitum-fed rats. These data are therefore consistent with reports that fasting increases plasma ghrelin (4), and they suggest that there was no further increase as a result of the weight loss. Although it might be argued that the anticipatory rise of ghrelin observed in the meal-fed rats is caused by the reduced body weight, we consider this unlikely because ad libitum-fed rats with normal body weight that were sampled before lights out also had an anticipatory increase of plasma ghrelin. There are therefore two factors that determine plasma ghrelin before a meal. The first is an increment caused by, and presumably proportional to, the period of food deprivation (5, 23), and the second is a larger increment that occurs when animals are able to anticipate eating. The difference between the two is even more striking considering that the ad libitum-fed rats most likely consumed the majority of their food during the first 4 h of the dark phase the night before the food was removed from their cages. Therefore, these rats consumed their last large meal more than 30 h before the beginning of the blood sampling. Keeping this in mind, plasma ghrelin levels would have been expected to be higher in the ad libitum-fed rats if the duration of food deprivation but not the anticipation of food were the driving factor for the ghrelin secretion. Similarly, the stress for the ad libitum-fed rats who were not used to an overnight food deprivation should have been greater, and ghrelin secretion was found to be increased during water immersion and restraint stress (24).
Collectively, this also speaks to the role of anticipation in the preprandial ghrelin peak observed in our meal-fed rats. One interpretation of the increased ghrelin response in the meal-fed rats is that the increase in circulating ghrelin contributes to the necessary metabolic adjustments the animals need to make to eat their full day’s ration of calories over a short period (16, 25) and that this anticipatory response becomes entrained over 14 d with a regimented meal pattern. Consistent with this interpretation, meal-fed rats eat significantly more food than ad libitum controls deprived for a comparable if not longer period of time. Another interpretation of these data is that both groups of rats were similarly hungry based on hours of deprivation, but the meal-fed group was able to eat more food than the ad libitum-fasted rats because of other meal-preparatory functions of ghrelin, beyond its orexigenic properties.
In meal-fed rats expecting their daily food, and in ad libitum-fed rats before lights out, plasma ghrelin increases to a peak around 30 min before the anticipated meal and then declines somewhat. This is consistent with previous results in meal-fed sheep (14). If there were a direct correspondence between plasma ghrelin and meal initiation, the ad libitum-fed rats should begin eating before lights out, at the time of the ghrelin peak, and in meal-fed rats, the peak should occur at the time food is expected rather than 30 min before. That they do not raises some question about the role of ghrelin in meal initiation. Furthermore, the fact that meal-fed animals eat steadily over the course of 4 h, consuming approximately 5 g in the first 15 min and approximately 19 g after 4 h despite a steep drop in plasma ghrelin, is also inconsistent with a purely orexigenic function. A more likely explanation is that the rise in circulating ghrelin is tied to the anticipation of feeding at a predictable time of day. In this setting, ghrelin may function to initiate preparatory responses that enable the animal to consume a large caloric load without significant disruption of nutrient homeostasis; previously demonstrated ghrelin effects on gastrointestinal or endocrine pancreatic function are potential sites of this action (26, 27, 28, 29, 30). This role would be analogous to, or possibly synergistic with, other anticipatory responses such as premeal, or cephalic, insulin secretion.
It is noteworthy that the preprandial ghrelin peak was considerably larger in meal-fed rats anticipating their daily 4-h period of food availability compared with ad libitum-fed rats before the onset of the dark. One interpretation of these data is that the higher ghrelin peak observed in the meal-fed rats is necessary to prepare them for the intake of a large amount of calories in a short window of time. Second, ad libitum-fed rats most likely start to eat at varying time points around the onset of the dark phase. Therefore, one could expect that their maximum plasma ghrelin level occurs with a larger temporal variance than occurs in the highly trained meal-fed rats that receive their food precisely at 1200 h every day. Although we have no direct evidence to rule out the possibility that the preprandial peak in plasma ghrelin in meal-fed rats was simply a strong orexigenic stimulus that elicits the increased food intake during their 4 h of food access, we do not consider this to be likely. Rather, we believe that the increment of ghrelin reflects the fact that more food will be consumed, and that it assists in preparing the body to receive the calories; however, additional studies, perhaps using GH secretagogue antagonists, will be necessary to gain additional insight into these questions.
Because eating large meals is a necessity of life in some environments, an important strategy for minimizing their impact is to initiate anticipatory responses before the meal that lessen the postprandial perturbation (reviewed in Refs.16 and 25). The best studied of these anticipatory responses is the premeal secretion of insulin to minimize the prandial increases in blood glucose (17). The phenomenon of cephalic insulin has been demonstrated in both rats (31) and humans (15, 32), and if cephalic insulin secretion is prevented or its action blocked, smaller meals are consumed (33). There are numerous other meal-anticipatory responses including a premeal reduction in metabolic rate, a premeal elevation of plasma corticosterone, and a premeal increase in body temperature (34, 35, 36). Interestingly, the preprandial rise of corticosterone is comparable to the increase in ghrelin. The involvement of both hormones in the regulation of food intake and energy balance warrants the exploration of the precise temporal relationship between these two hormones in future studies as well as how they may interact to prepare an individual for an imminent meal.
In addition to its effects on preprandial ghrelin, meal anticipation also had profound effects on postprandial ghrelin. In all previous studies to date, when food-deprived individuals ingest nutrients, plasma ghrelin drops rapidly (11, 13, 14); our data from experiment 4a in ad libitum-fed rats are consistent with this. However, when meal-fed rats and ad libitum-fed rats that were similarly food deprived subsequently ingested the same fixed amount of food, the postprandial ghrelin responses of these groups were dramatically different. Only the ad libitum deprived rats showed reduced ghrelin in response to their 5-ml meal of Ensure Plus, with plasma ghrelin returning to prefasting baseline values within 1 h. Meal-fed rats, on the other hand, after ingesting the same 5-ml meal showed no change of plasma ghrelin over the course of the 2 h assessed. The present findings suggest that anticipating a particular number of calories can influence the postprandial ghrelin response. In this study, meal-fed rats consumed only 12% of what they had been trained to expect over that time period and therefore did not suppress ghrelin; the identical number of calories completely suppressed ghrelin in ad libitum-fed rats deprived the same amount of time as meal-fed rats. Previous studies have shown that postprandial ghrelin suppression is directly proportional to calories ingested (11, 37). Our data are the first to demonstrate that postprandial ghrelin, similar to preprandial ghrelin, is dependent on meal anticipation as well as nutrient status.
Our data in rats appear to conflict with a report in humans suggesting that ghrelin functions as an appetite signal. In that study, plasma ghrelin levels rose before voluntary meal requests in subjects devoid of time or food cues, and this preprandial ghrelin rise was correlated with subjective hunger scores (38). However, it should be noted that the subjects in that study were accustomed to eating three meals per day, and given that they had been deprived of time cues for only a few hours, it is possible that their endogenous patterns of feeding behavior persisted. Thus, it is likely that anticipatory influences on ghrelin secretion are also present in humans, although this question will require rigorous examination.
In summary, the present results demonstrate that ghrelin secretion is regulated by learned anticipation independent of deprivation, in addition to regulation by nutrient status. Consistent with previous data from other species including humans, there is a learned, preprandial increase of plasma ghrelin when individuals have been trained to anticipate their food at a fixed time. When food deprivation was held constant, fasted ad libitum-fed rats maintained high, consistent ghrelin values over time, whereas meal-fed rats displayed a rise and a prominent peak of plasma ghrelin 30 min before food was expected. These data, when considered in the context of previous data, support a potential role for ghrelin in anticipatory meal preparation. Preprandial ghrelin may serve to prepare the stomach and gastrointestinal tract for anticipated food. Anticipation of food also had an impact on postprandial ghrelin, suggesting that the postprandial ghrelin response is not simply related to the number of calories ingested. Future studies are necessary to determine the mechanisms by which anticipation interacts with nutrient status to affect both pre- and postprandial ghrelin.
Acknowledgments
We thank Kay Ellis for her careful assay of plasma hormones and Kathi Smith, Joyce Sorrell, Eileen Elfers, Ken Parks, and Mouhamadoul Toure for technical assistance.
Footnotes
This work was supported by several grants from the National Institutes of Health: DK 63779 (D.L.D.), 17844 (S.C.W.), 56863 (D.A.D.), and 54080 (R.J.S.). The Obesity Research Center at the University of Cincinnati is supported in part by Procter & Gamble.
First Published Online September 22, 2005
Accepted for publication September 8, 2005.
References
Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660
Date Y, Kojima M, Hosoda H, Sawaguchi A, Mondal MS, Suganuma T, Matsukura S, Kangawa K, Nakazato M 2000 Ghrelin, a novel growth hormone-releasing acylated peptide, is synthesized in a distinct endocrine cell type in the gastrointestinal tracts of rats and humans. Endocrinology 141:4255–4261
Cummings DE, Shannon MH 2003 Roles for ghrelin in the regulation of appetite and body weight. Arch Surg 138:389–396
Tschop M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913
Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S 2001 A role for ghrelin in the central regulation of feeding. Nature 409:194–198
Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR 2001 Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992
Tolle V, Bassant MH, Zizzari P, Poindessous-Jazat F, Tomasetto C, Epelbaum J, Bluet-Pajot MT 2002 Ultradian rhythmicity of ghrelin secretion in relation with GH, feeding behavior, and sleep-wake patterns in rats. Endocrinology 143:1353–1361
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Fujimiya M, Katsuura G, Makino S, Fujino MA, Kasuga M 2001 A role of ghrelin in neuroendocrine and behavioral responses to stress in mice. Neuroendocrinology 74:143–147
Shintani M, Ogawa Y, Ebihara K, Aizawa-Abe M, Miyanaga F, Takaya K, Hayashi T, Inoue G, Hosoda K, Kojima M, Kangawa K, Nakao K 2001 Ghrelin, an endogenous growth hormone secretagogue, is a novel orexigenic peptide that antagonizes leptin action through the activation of hypothalamic neuropeptide Y/Y1 receptor pathway. Diabetes 50:227–232
Kamegai J, Tamura H, Shimizu T, Ishii S, Sugihara H, Wakabayashi I 2000 Central effect of ghrelin, an endogenous growth hormone secretagogue, on hypothalamic peptide gene expression. Endocrinology 141:4797–4800
Callahan HS, Cummings DE, Pepe MS, Breen PA, Matthys CC, Weigle DS 2004 Postprandial suppression of plasma ghrelin level is proportional to ingested caloric load but does not predict intermeal interval in humans. J Clin Endocrinol Metab 89:1319–1324
Masuda Y, Tanaka T, Inomata N, Ohnuma N, Tanaka S, Itoh Z, Hosoda H, Kojima M, Kangawa K 2000 Ghrelin stimulates gastric acid secretion and motility in rats. Biochem Biophys Res Commun 276:905–908
Cummings DE, Purnell JQ, Frayo RS, Schmidova K, Wisse BE, Weigle DS 2001 A preprandial rise in plasma ghrelin levels suggests a role in meal initiation in humans. Diabetes 50:1714–1719
Sugino T, Hasegawa Y, Kikkawa Y, Yamaura J, Yamagishi M, Kurose Y, Kojima M, Kangawa K, Terashima Y 2002 A transient ghrelin surge occurs just before feeding in a scheduled meal-fed sheep. Biochem Biophys Res Commun 295:255–260
Teff K 2000 Nutritional implications of the cephalic-phase reflexes: endocrine responses. Appetite 34:206–213
Woods SC 1991 The eating paradox: how we tolerate food. Psychol Rev 98:488–505
Berthoud H-R, Berieter DA, Trimble ER, Siegel EG, Jeanrenaud B 1981 Cephalic phase, reflex insulin secretion: neuroanatomical and physiological characterization. Diabetologia 20:393–401
Ensinck JW, Laschansky EC, Vogel RE, D’Alessio DA 1991 Effect of somatostatin-28 on dynamics of insulin secretion in perfused rat pancreas. Diabetes 40:1163–1169
Kissileff HR 1970 Free feeding in normal and "recovered lateral" rats monitored by a pellet-detecting eatometer. Physiol Behav 5:163–173
Strubbe JH, Keyser J, Dijkstra T, Prins AJ 1986 Interaction between circadian and caloric control of feeding behavior in the rat. Physiol Behav 36:489–493
Murakami N, Hayashida T, Kuroiwa T, Nakahara K, Ida T, Mondal MS, Nakazato M, Kojima M, Kangawa K 2002 Role for central ghrelin in food intake and secretion profile of stomach ghrelin in rats. J Endocrinol 174:283–288
Strubbe JH, Woods SC 2004 The timing of meals. Psychol Rev 111:128–141
Cummings DE, Weigle DS, Frayo RS, Breen PA, Ma MK, Dellinger EP, Purnell JQ 2002 Plasma ghrelin levels after diet-induced weight loss or gastric bypass surgery. N Engl J Med 346:1623–1630
Brzozowski T, Konturek PC, Konturek SJ, Kwiecien S, Drozdowicz D, Bielanski W, Pajdo R, Ptak A, Nikiforuk A, Pawlik WW, Hahn EG 2004 Exogenous and endogenous ghrelin in gastroprotection against stress-induced gastric damage. Regul Pept 120:39–51
Woods SC, Strubbe JH 1994 The psychobiology of meals. Psychonomic Bulletin and Review 1:141–155
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Ueno N, Makino S, Fujimiya M, Niijima A, Fujino MA, Kasuga M 2001 Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 120:337–345
Trudel L, Tomasetto C, Rio MC, Bouin M, Plourde V, Eberling P, Poitras P 2002 Ghrelin/motilin-related peptide is a potent prokinetic to reverse gastric postoperative ileus in rat. Am J Physiol Gastrointest Liver Physiol 282:G948–G952
Inui A, Asakawa A, Bowers CY, Mantovani G, Laviano A, Meguid MM, Fujimiya M 2004 Ghrelin, appetite, and gastric motility: the emerging role of the stomach as an endocrine organ. FASEB J 18:439–456
Lee HM, Wang G, Englander EW, Kojima M, Greeley Jr GH 2002 Ghrelin, a new gastrointestinal endocrine peptide that stimulates insulin secretion: enteric distribution, ontogeny, influence of endocrine, and dietary manipulations. Endocrinology 143:185–190
Reimer MK, Pacini G, Ahren B 2003 Dose-dependent inhibition by ghrelin of insulin secretion in the mouse. Endocrinology 144:916–921
Strubbe JH, Steffens AB 1975 Rapid insulin release after ingestion of a meal in the unanesthetized rat. Am J Physiol 229:1019–1022
Teff KL, Engelman K 1996 Palatability and dietary restraint: effect on cephalic phase insulin release in women. Physiol Behav 60:567–573
Louis-Sylvestre J 1976 Preabsorptive insulin release and hypoglycemia in rats. Am J Physiol 230:56–60
Rowe JW, Young JB, Minaker KL, Stevens AL, Pallotta J, Landsberg L 1981 Effect of insulin and glucose infusions on sympathetic nervous system activity in normal man. Diabetes 30:219–225
Powley TL, Berthoud HR 1985 Diet and cephalic phase insulin responses. Am J Clin Nutr 42:991–1002
de Vries J, Strubbe JH, Wildering WC, Gorter JA, Prins AJA 1993 Patterns of body temperature during feeding in rats under varying ambient temperatures. Physiol Behav 53:229–235
le Roux CW, Patterson M, Vincent RP, Hunt C, Ghatei MA, Bloom SR 2005 Postprandial plasma ghrelin is suppressed proportional to meal calorie content in normal-weight but not obese subjects. J Clin Endocrinol Metab 90:1068–1071
Cummings DE, Frayo RS, Marmonier C, Aubert R, Chapelot D 2004 Plasma ghrelin levels and hunger scores in humans initiating meals voluntarily without time- and food-related cues. The adipoinsular axis: effects of leptin on pancreatic -cells 287:E297–E304(Deborah L. Drazen, Torsten P. Vahl, Davi)