Eating Behavior and the Evolutionary Perspective on Anorexia Nervosa.

On the standard perspective, anorexia nervosa and other eating disorders are caused by genetically determined, neurochemically mediated mental illnesses. Standard treatment, cognitive behavioral therapy (CBT), targets cognitive processes thought to maintain the disorders. Effective neurochemically based treatments are not available and the rate of remission is ≤25% 1 year after CBT, with unknown outcomes in the long-term. With starvation as the major threat in biological history, the evolutionary perspective focuses on foraging for food and eating behavior. A neural network, including hypothalamic arcuate peptide-neurons, brainstem serotonin- and dopamine-neurons and their prefrontal cortical projections, mediates (rather than controls) the behavioral adaptations to variations in food availability; activation of the network is associated with opposing behavioral outcomes depending upon external variations. In the clinic, the control of eating behavior is therefore outsourced to a machine that provides feedback on how to eat. Hundreds of eating disorders patients have recovered by practicing eating; the rate of remission is 75% in on average 1 year of treatment, the rate of relapse is 10% over 5 years of follow-up and no patient has died. A two-parameter asymptotic exponential growth curve modeled the eating behavior of 17 healthy women but not that of 17 women with anorexia nervosa. When in remission, the eating behavior of the anorexic women approached that of the healthy women. It is suggested that the treatment of eating disorders should focus on eating behavior.

Cognitive behavior therapy for eating disorders versus normalization of eating behavior.

We examine the science and evidence supporting cognitive behavior therapy (CBT) for the treatment of bulimia nervosa and other eating disorders. Recent trials focusing on the abnormal cognitive and emotional aspects of bulimia have reported a remission rate of about 45%, and a relapse rate of about 30% within one year. However, an early CBT trial that emphasized the normalization of eating behavior had a better outcome than treatment that focused on cognitive intervention. In support of this finding, another treatment, that restores a normal eating behavior using mealtime feedback, has an estimated remission rate of about 75% and a relapse rate of about 10% over five years. Moreover, when eating behavior was normalized, cognitive and emotional abnormalities were resolved at remission without cognitive therapy. The critical aspect of the CBT treatment of bulimia nervosa therefore may actually have been the normalization of eating behavior.

Dopamine and anorexia nervosa.

We have suggested that reduced food intake increases the risk for anorexia nervosa by engaging mesolimbic dopamine neurons, thereby initially rewarding dieting. Recent fMRI studies have confirmed that dopamine neurons are activated in anorexia nervosa, but it is not clear whether this response is due to the disorder or to its resulting nutritional deficit. When the body senses the shortage of nutrients, it rapidly shifts behavior toward foraging for food as a normal physiological response and the mesolimbic dopamine neurons may be involved in that process. On the other hand, the altered dopamine status of anorexics has been suggested to result from a brain abnormality that underlies their complex emotional disorder. We suggest that the outcomes of the treatments that emerge from that perspective remain poor because they target the mental symptoms that are actually the consequences of the food deprivation that accompanies anorexia. On the other hand, a method that normalizes the disordered eating behavior of anorexics results in much better physiological, behavioral, and emotional outcomes.

Normalizing eating behavior reduces body weight and improves gastrointestinal hormonal secretion in obese adolescents.

HYPOTHESIS: Retraining obese adolescents to eat more slowly will lead to beneficial changes in circulating concentrations of gastrointestinal satiety hormones. METHODS: Ghrelin and peptide tyrosine-tyrosine were measured during an oral glucose tolerance test, at baseline and at 12 months during a randomized trial assessing the clinical effectiveness of a device (Mandometer) designed to retrain eating behavior. This computerized scale provided real-time feedback during meals in the intervention arm (n = 14) to slow down the speed of eating. The control group (n = 13) received only standard care aimed at improving lifestyle behavior. The Mandometer elicited greater improvements in weight loss than standard care. RESULTS: Compared with baseline, only those using the Mandometer exhibited lower mean levels of fasting ghrelin (48.14 ± 18.47 vs. 68.45 ± 17.78 pg/ml; P = 0.002) and mean ghrelin area under the curve (72.08 ± 24.11 vs. 125.50 ± 29.72 pg/ml × min; P < 0.001) at 12 months. Absolute mean suppression in ghrelin at 60 min was enhanced (-40.50 ± 21.06 vs. -12.14 ± 19.74 pg/ml × min; P = 0.001). Peptide tyrosine-tyrosine response at 90 min remained unaltered in the standard care arm, whereas those in the Mandometer arm increased (P < 0.001): the mean 90-min response increased by 72 pg/ml [95% confidence interval (CI) 52-92 pg/ml] between baseline and 12 months. In a partial correlation analysis adjusting for change (Δ) in body mass index sd scores, Δ meal duration correlated negatively with Δ absolute suppression in ghrelin at 60 min (r = -0.58; P = 0.037; 95% CI -0.79 to -0.27) and Δ ghrelin area under the curve (r = -0.62; P = 0.025; 95% CI -0.81 to -0.31). CONCLUSIONS: Retraining obese adolescents to eat more slowly has a significant impact on the gastrointestinal hormone response to a carbohydrate load, suggesting that externally modifiable eating behaviors actually regulate the hormonal response to food.

Obesity and the brain.

The world-wide increase in obesity has markedly stimulated research on the possibility that its cause can be found the brain. However, this research has produced little that can be used to treat obesity. The reason for the limited success of this approach may be that it relies on the hypothesis that the brain controls behavior. We suggest that this hypothesis is an artefact of the powerful tools used in behavioral neuroscience and that the brain has a permissive rather than causal role in eating behavior. Drugs affecting brain function are largely ineffective in treating obesity and may remain ineffective. Instead, we hypothesize that humans need external support to control body weight because they have evolved to pay a high physical price for food and are able to eat large amounts of food without constraints when that price is minimal. Two randomized controlled trials verify the hypothesis that support on how to eat normally and how to feel a normal level of fullness by use of on-line, real time feedback on a computer screen enables under- as well as overweight patients to adjust their eating behavior and improve their health.

How eating affects mood.

IOAKIMIDIS I, M. ZANDIAN, F. ULBL, C. BERGH, M LEON, AND P. SÖDERSTEN. How eating affects mood. PHYSIOL BEHAV 2011 (000) 000-000. We hypothesize that the changes in mood that are associated with eating disorders are caused by a change in eating behavior. When food is in short supply, the rhythm of the neural network for eating, including orbitofrontal cortex and brainstem, slows down and we suggest that this type of neural activity activates a partially overlapping neural network for mood, including dorsal raphe serotonin projections to the orbitofrontal and prefrontal cortex. As a consequence, people who restrict the amount of food that they consume, either by choice or by their limited access to food, become preoccupied with food and food-related behavior. Most eating disorders emerge from a history of dietary restriction and we suggest that disordered eating consequent upon food restriction produces the altered mental state of patients with eating disorders. Based on the present hypothesis, eating disorders are not the result of a primary mental disorder. Rather, this notion suggests that the patients should be treated by learning to eat an appropriate amount of food at an appropriate rate.

Behavioral neuroendocrinology and treatment of anorexia nervosa.

Outcome in anorexia nervosa remains poor and a new way of looking at this condition is therefore needed. To this aim, we review the effects of food restriction and starvation in humans. It is suggested that body weight remains stable and relatively low when the access to food requires a considerable amount of physical activity. In this condition, the human homeostatic phenotype, body fat content is also low and as a consequence, the synthesis and release of brain neurotransmitters are modified. As an example, the role of neuropeptide Y is analyzed in rat models of this state. It is suggested that the normal behavioral role of neuropeptide Y is to facilitate the search for food and switch attention from sexual stimuli to food. Descriptive neuroendocrine studies on patients with anorexia nervosa have not contributed to the management of the patients and the few studies in which hormones have been administered have, at best, reversed an endocrine consequence secondary to starvation. In a modified framework for understanding the etiology and treatment of anorexia nervosa it is suggested that the condition emerges because neural mechanisms of reward and attention are engaged. The neural neuropeptide Y receptor system may be involved in the maintenance of the behavior of eating disorder patients because the localization of these receptors overlaps with the neural systems engaged in cue-conditioned eating in limbic and cortical areas. The eating behavior of patients with anorexia nervosa, and other eating disorders as well, is viewed as a cause of the psychological changes of the patients. Patients are trained to re-learn normal eating habits using external support and as they do, their symptoms, including the psychological symptoms, dissolve.

Neuropeptide Y facilitates activity-based-anorexia.

The hypothesis that treatment with neuropeptide Y (NPY) can increase running activity and decrease food intake and body weight was tested. Female rats with a running wheel lost more weight than sedentary rats and ran progressively more as the availability of food was gradually reduced. When food was available for only 1h/day, the rats lost control over body weight. Correlatively, the level of NPY mRNA was increased in the hypothalamic arcuate nucleus. This phenomenon, activity-based-anorexia, was enhanced by intracerebroventricular infusion of NPY in rats which had food available during 2h/day. By contrast, NPY stimulated food intake but not wheel running in rats which had food available continuously. These findings are inconsistent with the prevailing theory of the role of the hypothalamus in the regulation of body weight according to which food intake is a homeostatic process controlled by "orexigenic" and "anorexigenic" neural networks. However, the finding that treatment with NPY, generally considered an "orexigen", can increase physical activity and decrease food intake and cause a loss of body weight is in line with the clinical observation that patients with anorexia nervosa are physically hyperactive and eat only little food despite having depleted body fat and up-regulated hypothalamic "orexigenic" peptides.

Psychoneuroendocrinology of anorexia nervosa.

It is suggested that the symptoms of anorexia nervosa are physiological responses to starvation. There is no evidence of a neural or non-neural dysfunction that predisposes women for anorexia nervosa and the endocrine and psychological consequences of starvation are reversed once patients have re-learnt how to eat and regained a normal body weight. Because variability in the supply of food may be a common evolutionary condition, it is more likely that body weight is variable than constant in normal circumstances. The role of the neuroendocrine system in times of feast and famine is to allow the individual to adopt behavioral strategies as needed rather than maintaining body weight homeostasis. Treatment of anorexic patients should aim at reducing their high level of physical activity in order to facilitate eating.

Intake inhibition by NPY and CCK-8: A challenge of the notion of NPY as an "Orexigen".

We tested the hypothesis that neuropeptide Y (NPY) interacts with cholecystokinin octapeptide (CCK-8) in inhibition of intake of an intraorally infused solution of sucrose, a test of consummatory ingestive behavior. Both intracerebroventricular infusion of NPY (10 microg) and intraperitoneal injection of CCK-8 (0.5 micro/kg) reduced the intake of a 1M solution of sucrose infused intraorally at a rate of 0.5 ml/min in ovariectomized female rats, but the two peptides did not interact in inhibiting intraoral intake. By contrast, NPY increased intake if the sucrose solution was ingested from a bottle, a test demanding both appetitive and consummatory ingestive responses. CCK-8 inhibited intake in this test and its inhibitory effect was increased by simultaneous treatment with NPY. The activity in the nucleus of the solitary tract (NTS), a brainstem relay mediating inhibition of intake, judged by the expression of c-fos-like immunoreactivity, was significantly increased after treatment with CCK-8 or NPY to approximately the same extent. Combined treatment with NPY and CCK-8 did not increase the c-fos-like immunoreactivity in the NTS above treatment with NPY or CCK-8 alone. These results strengthen the hypothesis that NPY, like CCK-8, is an inhibitor of consummatory ingestive behavior and suggest that this inhibition is mediated via the NTS.

Meal size, satiety and cholecystokinin in gastrectomized humans.

A wealth of data supports the idea that the stomach and cholecystokinin octapeptide (CCK-8) normally play important roles in meal size and satiety. We studied long-term gastrectomized humans to further evaluate this possibility. Ten humans, who were gastrectomized 8 (3-12) years earlier, and eight controls ate a meal from a plate placed on a scale connected to a computer and estimated their satiety every minute using a computerized rating scale. Blood levels of CCK-8 were measured before and after the meal. There was no difference between the groups in the amount of food consumed or in the perception of satiety during the meal. Gastrectomized humans had higher blood levels of CCK-8 than controls before the meal; the levels increased after the meal in the controls but not in the gastrectomized subjects. It is suggested that although the stomach and CCK-8 normally are involved in the control of meal size and satiety, their roles are dispensable.

Intake inhibition by NPY: role of appetitive ingestive behavior and aversion.

Intraventricular infusion of neuropeptide Y (NPY) decreases the amount female rats ingest during intraoral infusion (consummatory behavior) of a 1-M solution of sucrose at a rate of 0.5 ml/min and simultaneously increases the number of times the rats visit a bottle filled with sucrose (appetitive behavior). In this study, we investigated if the suppression of consummatory behavior was dependent upon the increase of appetitive behavior. The shift from consummatory to appetitive ingestive behavior was attenuated by adding 3-mM quinine HCl (QHCl) to the sucrose solution in the bottle. However, the intraoral intake of the sucrose solution was still decreased in NPY-treated rats. NPY did not modify taste reactivity as measured by aversive responses during continuous intraoral infusion of sucrose or ingestive and aversive responses to brief intraoral infusion of sucrose (0, 0.3 or 1 M) or QHCl (0, 0.3 or 3 mM). NPY stimulated visits to a bottle and intake from the bottle and inhibited sexual behavior in male rats but had no effect on the sexual behavior in the absence of a bottle. The visits and the intake were suppressed, but sexual behavior was not activated by adding QHCl (3 mM) to the solution in the bottle. Obstructing appetitive ingestive behavior, therefore, does not indiscriminately facilitate consummatory behavior. Male rats showed aversive or ingestive behavior and sexual behavior simultaneously during intraoral infusion of QHCl or condensed milk. It is suggested that NPY decreases intraoral intake and increases appetitive ingestive behavior via partially separable mechanisms that are independent of taste aversion.

Dopamine D(2) receptors and ingestive behavior: brainstem mediates inhibition of intraoral intake and accumbens mediates aversive taste behavior in male rats.

RATIONALE: One of the factors that terminate the ingestion of an intraorally infused solution of sucrose may be an increase in the perceived aversiveness of its taste. OBJECTIVES: We tested the hypothesis that dopamine D(2), as opposed to D(1), receptors in the brainstem or nucleus accumbens inhibit intraoral intake by enhancing the aversiveness of the taste of the infused solution. METHODS: Male rats were infused intraorally with a 2 M sucrose solution (1 ml/min) and intake and the display of gapes and chin rubs, i.e. taste-related aversive behavior, was measured. Gapes and chin rubs were also measured in rats during and 40 s after brief intraoral infusion (1 ml/min during 20 s) of a 0.3 mM solution of quinine HCl. The full D(1) receptor agonist dihydrexidine (0.1-3.0 mg/kg) and antagonist SCH-23390 (0.03-0.1 mg/kg), the D(2) receptor agonist quinpirole (0.3 mg/kg) and antagonist raclopride (1.7 mg/kg) were injected IP. Quinpirole (14-55 microg) and raclopride (5 microg) were also infused into the fourth brain ventricle. In addition, quinpirole (2 or 10 microg) was infused into the shell region of the nucleus accumbens. RESULTS: IP dihydrexidine and quinpirole inhibited the intraoral intake of sucrose and pretreatment with raclopride, but (in the case of dihydrexidine) not SCH-23390, attenuated this effect. Injection of quinpirole into the fourth ventricle produced raclopride-reversible inhibition of intraoral intake but did not stimulate the display of gapes and chin rubs. Infusion of quinpirole into the shell region of the nucleus accumbens had the opposite effects. The intake of sucrose was suppressed by the addition of quinine HCl but this suppression was unaffected by dopamine agonist or antagonist treatment. CONCLUSIONS: It is suggested that brainstem dopamine D(2) receptors mediate suppression of consummatory ingestive behavior and that D(2) receptors in the shell region of the nucleus accumbens mediate the display of gapes and chin rubs, but that neither of these D(2) receptor populations mediate the hedonic evaluation of taste.

Acute effects of peritoneal dialysis solutions on appetite in non-uremic rats.

BACKGROUND: Standard peritoneal dialysis (PD) solutions may contribute to anorexia in PD patients due to the peritoneal absorption of glucose from the dialysate, abdominal discomfort and other factors. New PD solutions containing alternative osmotic agents, neutral pH and bicarbonate as buffer were recently developed. To test the effect of these solutions on appetite, we investigated how intraoral (IO) intake of sucrose via an IO cannula was influenced by intraperitoneal (IP) infusion of different PD solutions in an appetite model in rats. METHODS: The IO intake was measured in male Wistar rats after an IP dwell of 30 and 120 minutes with the following PD solutions: 1.36%, 2.27% and 3.86% glucose based and lactate buffered solutions (D); 1.36%, 2.27% and 3.86% glucose based and bicarbonate/lactate buffered solutions (P); 7.5% icodextrin based solution (E); 1.1% amino acid-based solution (N); and, 2.5% glucose-based lactate-buffered solution (GB), using sham injection (injection without infusion) as control. Prior to the tests, rats were provided with an IO cannula, and were trained for two weeks until the rate of IO intake had stabilized. RESULTS: The D and N solutions inhibited IO intake. For the D solutions, the degree of appetite suppression was higher with the higher concentration of glucose. P 3.86%, but not P 1.36% and P 2.27% solutions, inhibited the IO intake. However, a comparison of the degree of appetite inhibition between D and P showed less inhibition with P 1.36%, 2.27% and 3.86% solutions than with corresponding D solutions. The E solution did not seem to suppress appetite. Finally, no significant difference in IO intake was found between rats given GB 2.5% and D 2.27%. CONCLUSIONS: In this appetite model in rats, the measurement of IO intake after the IP infusion of different dialysis solutions showed that (1) N and D solutions may reduce appetite, and for the D solutions the degree of appetite inhibition was related to the dialysate concentrations of glucose; (2) the P solutions had less impact on appetite than the D solutions; (3) the E solution had no impact on appetite during the short dwells of 30 and 120 minutes. The demonstrated differences between the different solutions appear to be due to different concentrations, and type, of nutrients used as osmotic agent (glucose, amino acids, icodextrin) or buffer (lactate), although differences in dialysate pH, tonicity and concentration of glucose degradation products also may be important. The present studies suggest a possible positive effect on appetite by using bicarbonate/lactate buffered solutions instead of lactate buffered solutions.

Locus coeruleus noradrenergic lesions attenuate intraoral intake.

I.p. injections of DSP-4 in male rats decreased norepinephrine (NE) levels to varying degrees throughout the brain with 66.7% reductions in the hypothalamic paraventricular nucleus. Intake of intraorally infused sucrose was reduced for 14 days but daily pellet intake recovered within 5 days post-injection. Intraventricular NE restored intraoral sucrose intake in DSP-4-lesioned rats without affecting controls. Intraventricular infusion of neuropeptide Y (NPY) reduced intraoral intake in controls but had no effect in DSP-4-lesioned rats. NPY markedly inhibited intraoral intake in DSP-4-treated rats that also received NE. These data confirm studies showing that NPY decreases consummatory ingestive behavior and suggest that this inhibition involves ascending noradrenergic projections from locus coeruleus.

Aversive behavior during intraoral intake in male rats.

We investigated if the taste of a sucrose solution becomes progressively more aversive during intraoral infusion and if this contributes to the termination of the intake in male rats. The display of aversive behavior, such as gapes and chin rubs, but not headshakes, forelimb flails or orofacial grooming, varied with the concentration of an intraorally infused solution of quinine hydrochloride (QHCl) and increased by the time the rat rejected an intraorally infused 2 M solution of sucrose. Activation of gapes and chin rubs by brief intraoral infusion of QHCl advanced the rejection of the sucrose solution if given late during intraoral infusion, but blockade of gaping by anaesthesization of the oral cavity with Xylocain did not prolong the intake of the sucrose solution. Headshakes and forelimb flails could be elicited by stimulating the head and limbs with sucrose, and gapes and chin rubs were activated by infusion of a 2 M solution of sucrose into the stomach or duodenum but not by infusion of glucose into the jugular or hepatic portal vein. Preventing filling of the gastrointestinal tract during intraoral infusion of sucrose (sham feeding) eliminated the display of gapes and chin rubs. It is suggested that an increase in the aversiveness of the taste of a sucrose solution contributes to the rejection of that solution during intraoral infusion. However, rats can reject a sucrose solution in the absence of any behavioral sign of aversion and none of the so-called "taste-related" aversive behaviors is exclusively dependent upon stimulation of the taste receptors in the oral cavity.

Evidence that MK-801 stimulates intraoral intake by acting on hepatic afferents.

Satiety signals from the gastrointestinal tract travel via vagal afferents to the nucleus of the solitary tract (NTS) in the brain stem, the first central relay in a neural network which controls food intake. The non-competitive NMDA antagonist MK-801 facilitates food intake in rats by acting on the NTS. Here we report that hepatic portal vein infusion of MK-801 (25 or 50 microg/kg) increases intake of an intraorally infused 1 M solution of sucrose (by 113 +/- 9 and 132 +/- 11%, respectively) and that this effect is prevented by hepatic vagotomy. By contrast, jugular vein infusion of MK-801 fails to increase sucrose intake but induces forward locomotion, indicating activation of a central mechanism. These data suggest that MK-801 can stimulate food intake by acting peripherally on hepatic vagal afferents.

NPY-leptin: opposing effects on appetitive and consummatory ingestive behavior and sexual behavior.

Many studies have indicated that neuropeptide Y (NPY) stimulates and leptin inhibits food intake. In line with this, intracerebroventricular injection of NPY (10 microg) stimulated and leptin (10 microg) inhibited intake of a sucrose solution when female rats were required to obtain the solution from a bottle. However, NPY inhibited and leptin stimulated intake if the solution was infused intraorally. Thus NPY stimulates the responses used to obtain food but inhibits those used to consume food, and leptin has the opposite effects. To test the specificity of these responses the sexual behavior of male rats was examined. NPY-treated males showed minor deficits in sexual behavior but chose to ingest a sucrose solution rather than copulate with a female if offered the choice. By contrast, leptin-treated males ingested little sucrose and displayed an increase in ejaculatory frequency if given the same choice. It is suggested that NPY is not merely an orexigenic peptide, but one that directs attention toward food. Similarly, leptin may not be an anorexic peptide, but one that diverts attention away from food toward alternate stimuli.