There was a strong correspondence between the GIP secretory response (Fig. 2, right) and the magnitude and time course of reported fullness in response to manipulation of meal size (Fig. 2, left), suggesting the important role of nutrient transit through the gastrointestinal tract on this measure of short-term satiation.
In contrast to the insensitivity of appetite response, the putative appetite controlling hormones responded appropriately to negative morning energy balance caused by either a small meal or exercise energy expenditure: ghrelin with increases (Fig. 3, left), leptin with decreases (Fig. 3, right), and both to correction of energy imbalance via TPN infusions.
The failure of appetite to respond to short-term fluctuations in energy balance was paralleled by the failure of these fluctuations to produce increased food consumption during the midday meal (Fig. 4). As was the case in studies manipulating energy density of food (15), appetite and meal-to-meal eating seemed to be controlled by stomach capacity rather than by preceding energy deficit. In the absence of sustained energy restriction or weight loss, human appetite and meal-to-meal eating were unresponsive to short-term variations in energy availability caused by inadequate meal size, energy expenditure of exercise, or intravenous increases in nutrient energy.
REINTERPRETATION OF THE ENERGY REGULATORY MECHANISM
Collectively, our research (coupled with other lines of evidence on the opportunistic responses of human appetite to environmental circumstances and its unresponsiveness to short-term variations in energy availability) support the concept of appetite's nonhomeostatic control when not constrained by food restriction or a substantial weight loss. In fundamental design, a human's meal-to-meal consumption pattern does not differ from that of a blowfly. When a blowfly steps into a sugar solution, its taste receptors on the foot pads trigger ingestion of sweet fluid in proportion to its concentration and palatability. A full crop inhibits further feeding through a message sent by a recurrent nerve to the brain, and avulsion of this negative feedback from the crop leads to overeating (7).
Nonhomeostatic Contribution of Physical Activity to Energy Regulation
According to the current view of energy regulation, increases in body fat level should increase adiposity hormone negative feedback on feeding and facilitate physical activity and other forms of energy expenditure to produce a loss of calories and body fat. Conversely, decreases in body fat level should reduce adiposity hormone negative feedback over feeding and suppress catabolic processes that include physical activity and metabolic energy expenditure to restore energy balance (25). Injections of leptin into leptin-deficient mice that are obese and profoundly hypoactive seem to support this concept by stimulating levels of physical activity (22).
Physical activity can produce weight losses and reduce the level at which body fat is regulated in both rodents (27) and humans (13), regardless of the type of prevailing maintenance diets, when it is done either spontaneously or enforced. In addition, during recovery from enforced weight loss, exposure to forced exercise results in a lower recovery weight plateau than does ad libitum eating without exercise (20). Physical activity may contribute to lowering weight plateaus in humans and rodents, in part, by suppressing appetite (Fig. 1) and preventing increases in feeding in compensation for exercise energy expenditure (Fig. 4).
The question of whether these effects of physical activity are homeostatic should be examined in the context of manipulations that affect energy balance. Much evidence overwhelmingly fails to support a homeostatic role of physical activity in energy regulation. There is a strong negative relationship between spontaneous physical activity and body fat in both animals and humans (Fig. 5). Obesity in rodents is invariably associated with profound hypoactivity whether induced by high-fat diet feeding (27), lesions of medial basal hypothalamus (16), or the absence or inactivation of leptin or its receptors (22). Morbidly obese humans also are reported to be almost completely inactive (32). Conversely, maintaining rats on restricted access to food leads to 300%-500% increases in spontaneous running as their weight loss increases. If the experiment is not terminated, rats virtually run themselves to death from energy depletion (24). This animal model is paralleled by human anorexia nervosa, a condition of suppressed food intake, sustained weight loss, and high motivation for, and involvement in, physical activity (6). Thus, the prevailing evidence indicates that spontaneous physical activity levels are related to body fat in an inverse and nonhomeostatic fashion.
Leptin and Insulin Act on Brain Substrates of Reward to Mediate Nonhomeostatic Eating and Locomotion
A way of reconciling the conflicting evidence regarding the restraining effects of insulin and leptin on appetite and body weight and fat gain (25) with the evidence that they do not influence meal-to-meal eating (4) is to consider the actions of the two hormones on the brain substrates of reward. Evidence links actions of insulin and leptin to spontaneous food seeking and locomotor behaviors through their suppression of the relevant brain substrates of reward. Insulin secretion accompanies ingestion of food, and the magnitude of this response is related to size and composition of the meal. Increases in insulin concentration associated with meal eating suppress both motivation to eat (26) and the motivation for physical activity. The well-known behavioral sequence in animals (and perhaps many humans) is to display somnolence and engage in sleep after a meal. Direct suppression by insulin injections of spontaneous physical activity in rats is seen at plasma concentrations that are too low to cause hypoglycemia and hunger (5). Similarly, hyperinsulinemia, that accompanies either lesions of VMH (9) or overeating after enforced weight loss (20), is accompanied by reductions in the levels of spontaneous physical activity or motivation to run on a treadmill. When hyperinsulinemia after the VMH lesions is prevented with subdiaphragmatic vagotomy, hyperphagia, obesity, and hypoactivity do not develop (9). Increased insulin secretion during the absorptive period stimulates increased transcription of the leptin gene. Leptin, in turn, contributes to suppression of locomotor behavior as is shown for hyperactivity in anorexic subjects (8) and semistarved rats (10). Thus, the absorptive increases in insulin and leptin contribute to postmeal suppression of behavioral activation.
Absorption of a meal and decline of insulin to its basal levels also are associated with a decline in plasma leptin concentrations (Fig. 3, right). As concentrations of both hormones decline, behavioral activation and spontaneous physical activity increase (14). Thus, declines in insulin and leptin secretion after digestion and absorption of a meal increase the motivation for locomotor behavior. This implicates the two hormones at their low postabsorptive concentrations in facilitation of nonhomeostatic meal-to-meal feeding and stimulation of nonhomeostatic spontaneous locomotor activity.
In view of exercise-associated appetite suppression and increases in lipolysis, it would be logical to expect that these effects would be mediated by increased leptin release during exercise. This inference to date has not been substantiated, although several studies report exercise-associated increases in plasma leptin concentration (12,18). Moderate exercise-associated leptin increases also are noticeable in our study (Fig. 3, right) and coincide with temporary suppression of hunger (Fig. 1). If more definitive support is found for the role of leptin in exercise-induced appetite suppression, this leptin action could serve as evidence for its role in functional linkage of meal eating and behavioral activation, and it could reveal a mechanism through which postmeal restraint over feeding contributes to a reduction in the weight plateau that is observed in exercising animals and humans.
Substantial evidence suggests that involvement of insulin and leptin in nonhomeostatic control of meal eating and physical activity is mediated through their actions on the brain substrates of reward. Two hypothalamic circuits serving motivation and reward have been described (2). One responds to fluctuation in short-term energy availability and body fat levels (and, for convenience, will be called homeostatic) and the other one is activated in the obese condition and preferentially responds to external sensory aspects of food (hedonic circuit). The homeostatic circuit comprises a large area of the medial shell of the nucleus accumbens, the ventral pallidum, the medial part of the lateral hypothalamus, and the ventral tegmentum. It is activated by μ opioid, cannabinoid CB 1, and dopamine receptor stimulation. Such stimulation increases the drive to eat, obtain addictive rewards, and locomote. Both insulin and leptin receptors are present in, and are shown to influence, the homeostatic brain circuit of reward (11). Low insulin and leptin levels enhance, whereas high insulin and leptin levels dampen, the rewarding value of stimulation of this circuit by suppressing the release of dopamine and related neurotransmitters (14).
How Nonhomeostatic Meal-to-Meal Eating and Locomotion Help Stabilize Different Weight Plateaus
The evidence that meal-to-meal eating (4) and spontaneous physical activity operate in a nonhomeostatic fashion leaves open the question regarding how stable weight plateau that range from extreme leanness to morbid obesity are established and maintained. The resolution may reside in the roles of insulin and leptin within the temporal organization of feeding and behavioral activation during a 24-h day. This organization involves a circadian pattern of feeding and behavioral activation during approximately one half of the day and behavioral quiescence and suppression of feeding during the other half. The circadian pattern to human hunger ratings is evident from Figure 6. In this study, 10 healthy postmenopausal women engaged in a sedentary trial and in two exercise trials requiring two bouts of treadmill walking differing in intensity but producing a similar workload (35). Two aspects of the circadian appetite organization are seen.
First, hunger ratings are lower at the transition between waking and nocturnal phases of the day and are absent during the night. Second, hunger ratings are equally high at habitual feeding times in the sedentary and exercise trials and unaffected by the energy cost of exercise. The same was true in the previously described study despite variations in energy availability caused by exercise, different size meals, or intravenous nutrient infusions (Fig. 1). Meal-associated ratings of hunger show an ultradian (<24 h) periodicity approximately the duration of time it takes to digest and absorb a meal, and at such times the hunger ratings are refractory to changes in energy availability. This suggests a presence of a nonhomeostatic coordinating central nervous influence that activates hunger at habitual meal times in an all-or-none fashion.
No feeding mechanism can effectively support survival without a capacity to respond to energy depletion. A functional connection between the nonhomeostatic control of meal-to-meal appetite and eating, and of spontaneous locomotor activity, resides in the sensitivity of the neural substrates of reward to variations in energy availability and levels of body fat. The reward valuation in the homeostatic drive circuit that motivates food seeking, hypothalamic self-stimulation, and locomotion is potentiated by food deprivation and body fat losses (2). This is accomplished, in part, by decreases in the inhibitory effects of circulating insulin and leptin on the neural substrates of reward when they reach low levels during negative energy balance and body fat loss. Therefore, increased hunger and behavioral activation can be viewed as mechanisms that operate only when the size and caloric content of meals fall short of producing increased plasma insulin and leptin between habitual feeding times. If the ingested meal is small, as was the case with the 100-kcal morning meal in our study, ratings of hunger are increased between the habitual meal times (Fig. 1), and behavior activation also is expected to be higher. The greater the shortfall in meal energy or weight loss and concomitant decline in insulin and leptin concentrations, the greater the activation of the homeostatic substrates of reward and increased motivation to seek food and locomote (Fig. 7). A lifestyle requiring a substantial amount of physical work to procure food of moderate or low nutrient density will sustain motivation for both food seeking and physical activity and lead to a lean body weight plateau.
On the other hand, easy access to palatable and energy-rich food without the necessity for substantial amounts of physical work will favor excess caloric intake and progressive weight gain. In step with increases in body fat, basal concentrations of leptin and ghrelin will rise and inhibit locomotor behavior. The hedonic brain circuit, uncoupled from the state of body energy balance, will ensure intake of palatable food. In such an environment, the weight will stabilize at an overweight or obese plateau, supported by relative inactivity and sustained intake of palatable food.
Metabolic Contributions of Insulin and Leptin to Maintenance of Variable Weight Plateaus
Epidemic increases in human obesity and the ease of producing dietary obesity in animals may be attributed in part to the hedonic brain circuit of reward (2). This circuit comprises a smaller area of nucleus accumbens and ventral pallidum, orbitofrontal, anterior cingulate, and insular cortex; hippocampus and amygdala in the limbic forebrain; and mesolimbic dopamine projections to the nucleus accumbens and cortical sites. As with the homeostatic circuit, a hedonic pathway is activated by dopamine, μ opioids, and endocannabinoids. The hedonic circuit typically does not respond to declines in energy balance but is engaged at stable body weight plateaus in association with body fat increases and obesity, and is responsible for heightened finickiness to taste of food. Obese animals overeat palatable food, but lose weight and undereat unpalatable or bland foods (27). Both brain substrates of reward thus contribute to acquisition and storage of nutrient energy, the homeostatic one by increasing appetite and locomotion in response to energy shortage, and the hedonic one by facilitating seeking palatable food uncoupled from energy shortage.
Leptin and insulin also contribute to stability of body weight and body fat levels through their metabolic actions. Insulin resistance develops at high body fat levels and limits further nutrient storage and fat synthesis by blunting antilipolytic and lipogenic insulin actions. In this way, high basal plasma insulin concentrations contribute to maintenance of obese body fat levels. During food restriction and body weight losses, higher sensitivity of adipose and other peripheral tissues to insulin favors glycogen and lipid synthesis and drives the restoration of higher body fat levels.
The magnitude of a nocturnal rise in leptin concentration is proportional to the nutritional state at the onset of darkness. Nocturnal leptin concentration and presumably its lipolytic action (33) are greater in response to energy surplus and lower after energy depletion (31). Dawn and dusk nadirs of human hunger (Fig. 6) may reflect circadian transition from higher nocturnal leptin concentrations when lipolysis is increased and behavioral activation blocked to the lower diurnal leptin concentrations when leptin may affect the motivation to locomote in conjunction with meal eating. When humans deliberately truncate the period of nocturnal rest, declines in plasma leptin and increases in plasma ghrelin are associated with increased appetite and presumably greater fat synthesis (29). However, as levels of obesity rise and tissue resistance to leptin action increases (26), nocturnal lipolytic effects of leptin would be expected to decline and tend to maintain higher body fat levels.
The absence of a controlling role for leptin and insulin in meal-to-meal eating prompts a reinterpretation of their roles in energy regulation. Our research and other evidence indicate that meal-to-meal feeding, appetite, and spontaneous physical activity operate in a nonhomeostatic fashion, whereas plasma leptin and insulin respond to short-term fluctuations in energy availability and bear no relationship to human appetite. The nonhomeostatic character of meal-to-meal feeding and spontaneous physical activity is mediated by inhibitory actions of insulin and leptin on the brain substrates of reward. After meals and in response to body fat gain, increases in the concentration of these two hormones reduce the incentive value of food and behavioral activation. Negative energy balance and body fat losses reduce basal concentrations of both hormones and increase the motivation to seek food and locomote. A circadian clock produces maximal facilitation of appetite at habitual meal times, regardless of the prevailing energy balance, and thus contributes to a nonhomeostatic pattern of feeding. The contribution of both hormones to weight stability is mediated by their modulation of the responsiveness of brain substrates of reward to changes in momentary energy availability and body fat level and by producing metabolic effects that favor retention of higher body energy stores. The responsiveness of brain substrates of reward to reduced energy availability and body fat losses favors increases in food seeking motivation and behavioral activation and drives weight and fat gain after body fat losses. The homeostatic brain substrate of reward becomes refractory at high body fat levels when resistance to both hormones develops. In addition, a hedonic brain circuit that is uncoupled from responsiveness to energy deficit favors opportunistic ingestion of palatable food.
Human feeding behavior is thus designed for nonhomeostatic meal-to-meal eating and is influenced by social forces that interfere with optimal operation of the energy regulatory mechanism. Environmental circumstances and the design of controls of eating behavior enhance the motivation for food seeking when energy deprived and after body fat loss but also encourage humans to overeat highly palatable and energy-rich foods when not energy deprived. This suppresses physical activity and engages the hedonic brain circuit, both of which contribute to excess fat synthesis and accumulation. Our currently inadequate understanding of the energy regulatory mechanism limits the opportunity to adjust conditions that would allow optimal operation of feeding behavior and physical activity for maintenance of healthy weight. The need for additional knowledge is particularly acute in four areas:
1. definition of quantities and palatability characteristics of human meals that would preserve the ultradian motivation to be physically active (Fig. 5);
2. characterization of types, volumes, and intensity of physical activity that would produce optimal diurnal between-meal appetite suppression and nocturnal lipolysis;
3. coordinated research on the role of the hormones that track the changes in energy balance (e.g., leptin, insulin, ghrelin, and gut peptides) by interacting with the neurochemical and neuroanatomical control of feeding and physical activity in the context of ultradian and nyctohemeral organization; and
4. assessment of the feasibility of restructuring the human environment to reduce exposure to highly palatable and energy-rich food and increase the opportunity for physical activity in the workplace (30) and elsewhere.
Because of the journal limit on the number of references, the author regrets her inability to cite the work of many authors deserving recognition.
This study was supported by National Institutes of Health grants R15 DK066286 to K.T.B., DK20572 to Michigan Diabetes Research and Training Center, and M01-RR00042 to Michigan Clinical Research Unit.
1. Adolph EF. Urges to eat and drink in rats. Am. J. Physiol
. 1947; 151:110-25.
2. Berridge KC. "Liking" and "wanting" food rewards: Brain substrates and roles in eating disorders. Physiol. Behav
. 2009; 97:537-50.
3. Black AE, Coward WA, Cole TJ, Prentice AM. Human energy expenditure in affluent societies: An analysis of 574 doubly-labelled water measurements. Eur. J. Clin. Nutr
. 1996; 50:72-92.
4. Borer KT, Wuorinen E, Ku K, Burant C. Appetite responds to changes in meal content, whereas ghrelin, leptin, and insulin track changes in energy availability. J. Clin. Endocrinol. Metab
. 2009; 94:2290-8.
5. Campbell BA, Fibiger HC. Effects of insulin on spontaneous activity during food deprivation. J. Comp. Physiol. Psychol
. 1970; 71:341-6.
6. Casper RC. The "drive for activity" and "restlessness" in anorexia nervosa: Potential pathways. J. Affect. Disord
. 2006; 92:99-107.
7. Dethier VG, Solomon RL, Turner LH. Sensory input and central excitation and inhibition in the blowfly. J. Comp. Physiol. Psychol
. 1965; 60:303-13.
8. Ehrlich S, Burghardt R, Schneider N, et al
. The role of leptin and cortisol in hyperactivity in patients with acute and weight-recovered anorexia nervosa. Prog Neuropsychopharmacol. Biol. Psychiatry
9. Eng R, Gold RM, Sawchenko PE. Hypothalamic hypoactivity prevented but not reversed by subdiaphragmatic vagotomy. Physiol. Behav
. 1978; 20:637-41.
10. Exner C, Hebebrand J, Remschmidt H, et al
. Leptin suppresses semi-starvation induced hyperactivity in rats: Implications for anorexia nervosa. Mol. Psychiatry
11. Figlewicz DP, Benoit SC. Insulin, leptin, and food reward: Update 2008. Am. J. Physiol. Regul. Integr. Comp. Physiol
. 2009; 296:R9-R19.
12. Fisher JS, Van Pelt RE, Zinder O, Landt M, Kohrt WM. Acute exercise effect on postabsorptive serum leptin. J. Appl. Physiol
. 2001; 91:680-6.
13. Franz MJ, VanWormer JJ, Crain L, et al
. Weight-loss outcomes: A systematic review and meta-analysis of weight-loss clinical trials with a minimum 1-year follow-up. J. Am. Diet Assoc
. 2007; 107:1755-67.
14. Fulton S, Woodside B, Shizgal P. Modulation of brain reward circuitry by leptin. Science
15. Kendall A, Levitsky DA, Strupp BJ, Lissner L. Weight loss on a low-fat diet: Consequence of the imprecision of the control of food intake in humans. Am. J. Clin. Nutr
. 1991; 53:1124-9.
16. Kennedy GC, Mitra J. Hypothalamic control of energy balance and the reproductive cycle in the rat. J Physiol
. 1963; 166:395-407.
17. King NA, Burley VJ, Blundell JE. Exercise-induced suppression of appetite: Effects on food intake and implications for energy balance. Eur. J. Clin. Nutr
. 1994; 48:715-24.
18. Kraemer RR, Durand RJ, Acevedo EO, et al
. Effects of high-intensity exercise on leptin and testosterone concentrations in well-trained males. Endocrine
19. Levitsky DA. The non-regulation of food intake in humans: Hope of reversing the epidemic of obesity. Physiol. Behav
. 2005; 86:623-32.
20. MacLean PS, Higgins JA, Wyatt HR, et al
. Regular exercise attenuates the metabolic drive to regain weight after long-term weight loss. Am. J. Physiol. Regul. Integr. Comp. Physiol
. 2009; 297:R793-R802.
21. Morton GJ, Cummings DE, Baskin DG, Barsh GS, Schwartz MW. Central nervous system control of food intake and body weight. Nature
22. Pelleymounter MA, Cullen MJ, Baker MB, et al
. Effects of the obese gene product on body weight regulation in ob/ob mice. Science
23. Raynor HA, Epstein LH. Dietary variety, energy regulation, and obesity. Psychol. Bull
. 2001; 127:325-41.
24. Routtenberg A, Kuznesof AW. Self-starvation of rats living in activity wheels on a restricted feeding schedule. J. Comp. Physiol Psychol
. 1967; 64:414-21.
25. Schwartz MW, Woods SC, Porte DJ, Seeley RJ, Baskin DG. Central nervous system control of food intake. Nature
26. Schwartz MW, Woods SC, Seeley RJ, Barsh GS, Baskin DG, Leibel RL. Is the energy homeostasis system inherently biased toward weight gain? Diabetes
27. Sclafani A, Springer D. Dietary obesity in adult rats: Similarities to hypothalamic and human obesity syndromes. Physiol. Behav
. 1976; 17:461-71.
28. Shepard TY, Weil KM, Sharp TA, et al
. Occasional physical inactivity combined with a high-fat diet may be important in the development and maintenance of obesity in human subjects. Am. J. Clin. Nutr
. 2001; 73:703-8.
29. Spiegel K, Knutson K, Leproult R, Tasali E, Van Cauter E. Sleep loss: A novel risk factor for insulin resistance and Type 2 diabetes. J. Appl. Physiol
. 2005; 99:2008-19.
30. Thompson WG, Foster RC, Eide DS, Levine JA. Feasibility of a walking workstation to increase daily walking. Br. J. Sports Med
. 2008; 42:225-8.
31. van Aggel-Leijssen DP, van Baak MA, Tenenbaum R, Campfield LA, Saris WH. Regulation of average 24 h human plasma leptin level; the influence of exercise and physiological changes in energy balance. Int. J. Obes. Relat. Metab. Disord
. 1999; 23:151-8.
32. Vanhecke TE, Frankin BA, Miller WM, deJong AT, Coleman CJ, McCullough PA. Cardiorespiratory fitness and sedentary lifestyle in the morbidly obese. Clin. Cardiol
. 2009; 32:121-4.
33. Wang MY, Leem Y, Unger RH. Novel form of lipolysis induced by leptin. J. Biol. Chem
. 1999; 274:17541-4.
34. Welch I, Saunders K, Read NW. Effect of ileal and intravenous infusions of fat emulsions on feeding and satiety in human volunteers. Gastroeneterology
35. Wuorinen E. Detection of Exercise Energy Expenditure
. Ph.D. Dissertation, University of Michigan, Ann Arbor, MI; 2007.
This article has been cited
Keywords:©2010 The American College of Sports Medicine
leptin; insulin; ghrelin; motivation; reward; obesity