Neurobiological Clues to Anorexia Nervosa

There are increasing attempts to understand the biological underpinnings of eating disorders. Animal models of disordered eating are one potential avenue for this research. One approach is the activity-based anorexia model, in which food-restricted rodents are allowed access to a running wheel. The rodents develop severe weight loss and exercise excessively (Klenotich et al, 2012). This has been a useful model, but it differs in some respects from anorexia nervosa (AN) in humans. A good example is the prominence of exercise.

A new mouse model has been described that may be closer to the AN seen in humans. In this model, mice with a variant in the brain-derived neurotrophic factor (BDNF) gene were studied (Transl Psychiatry. 2016; 6:1). This BDNF gene variant has been tied to increased risk for AN, increased anxiety in humans, and anxiety-like symptoms in mice, and may be important for explaining risk for anxiety and depression in adolescents who are exposed to stress.

In this study, female mice with or without the BDNF variant were subjected to a social isolation stress at the age of 5 weeks (the equivalent of adolescence in mice). In some, calorie intake was restricted at 7 weeks of age. In response to this combination of genetic loading, social stress, and modest caloric restriction, approximately 40% of the mice developed severe caloric restriction, which appeared to be lasting. The effects occurred in both female and male mice, but appeared to be stronger in females.

Several aspects of this model (the preferential effects in female mice, the use of a social stress exposure, and modest caloric restriction, as well as the onset of severe and enduring food restriction) closely mirror what we see in patients with AN. This laboratory model may become a very useful tool for studying AN.