A preclinical model of binge eating elicited by yo-yo dieting and stressful exposure to highly palatable food

Preclinical models are needed to investigate the neurobiology and psychobiology of binge eating (BE) and to identify innovative pharmacotherapeutic strategies. BE episodes are characterized by uncontrollable, distressing eating of a large amount of highly palatable food (HPF). These episodes represent a central feature of bingeing-related eating disorders, such as binge eating disorder, bulimia nervosa, and binge/purge subtype anorexia nervosa. In line with the hypothesis that dieting and stress are key etiological determinants of BE the proposed model combines cycles of food restriction–refeeding and acute stress to evoke BE for sweet HPF. Four groups of female rats were employed: group 1 was normally fed and not stressed on the test day (25th); group 2 was fed normally but was exposed to an acute stress on day 25; group 3 was exposed to three cycles (4 days 66% of chow intake+4 days food ad libitum) of yo-yo dieting but not stressed; and group 4 was exposed to cyclic yo-yo dieting and then stressed. All groups were fed highly palatable food (HPF) for 2 h on days 5–6 and 13–14. Acute stress was elicited by exposing rats to HPF, but preventing them from access to it for 15 min, even though they were able to see it and to smell its odor. The combination of cyclic food restriction and stressful exposure to food markedly increased HPF intake. To test the predictive validity of this model, three drugs were evaluated: sibutramine, fluoxetine, and topiramate, chosen since clinical studies have reported that they may be effective in the treatment of bingeing-related eating disorders. Topiramate selectively inhibited compulsive HPF intake in rats submitted to caloric restriction and stress, thus, its effect was evident only in conditions in which binge-type eating occurred. Sibutramine and fluoxetine inhibited food intake in all conditions suggesting that their effect is a general effect on appetite and/or satiety and is not selective on the binge-type behavior. Midazolam was included to assess whether an anxiolytic benzodiazepine might reduce BE in response to a stressful procedure. Midazolam did not modify binge eating BE rats, but increased HPF intake in the other three groups of rats, as previously observed in animals and humans. Pharmacological results suggest that this model, in addition to face validity as an isomorphic model of human binge eating, is endowed with good predictive validity. Thus, the present model may be useful to evaluate new pharmacological strategies for the treatment of bingeing related eating disorders.