Testing Treatments in the Anorexic Mouse

800px-treadmills_at_gymAnorexia nervosa is one of the most common psychiatric disorders, and also the deadliest. But unlike depression, anxiety, or schizophrenia, no drugs are approved for the treatment of patients with this eating disorder. Non-pharmacological interventions, such as family-based treatment or cognitive-behavioral therapy, remain the gold standard for helping patients get back to healthy eating habits and weight levels. Occasionally, clinicians treating anorexia prescribe off-label use of drugs originally intended for other psychiatric disorders, but data is lacking on the effectiveness of those agents against this disease.

One obstacle against the development of drugs to treat anorexia nervosa has been a shortage of laboratory research into the disorder. For most diseases, drugs are thoroughly tested in animal models before reaching the clinic. But in anorexia nervosa, only a handful of studies have tried to untangle the pharmacology of effective treatment.

“Anorexia nervosa is the most deadly psychiatric disorder, and yet no approved pharmacological treatments exist,” said Stephanie Dulawa, assistant professor of Psychiatry & Behavioral Neuroscience at the University of Chicago Medicine and senior author of the study. “So one wonders why there isn’t more basic science work being done to better understand the mechanisms and to identify novel pharmacological treatments.”

In response to this research void, Dulawa’s laboratory, which models psychiatric conditions such as obsessive-compulsive disorder in mice, forged a collaboration with Daniel Le Grange, director of the Eating Disorders Program at the University of Chicago Medicine. In a recent paper published in the journal Neuropsychopharmacology, first author Stephanie Klenotich adapted an animal model of the eating disorder called activity-based anorexia (ABA) for mice, and took the first steps toward designing a better drug to help patients battle their disease.

In the ABA model, a combination of exercise and food restriction produces behaviors that are remarkably similar to anorexia nervosa in humans. Mice or rats are given 24-hour access to a running wheel in their cage and only allowed access to food for a finite amount of time each day instead of the limitless supply normally given to laboratory animals. After a time, the rodents will paradoxically exercise more and eat less, leading to severe weight loss that can be fatal without intervention. The model was originally developed in rats, but Klenotich wanted to adapt ABA to mice to allow for more sophisticated genetic manipulation. She found that a particular strain of mouse called Balb/cJ and six hours of food access produced ABA behavior that would be sensitive enough to test drug effectiveness.

For the first experiments, Le Grange tipped off his collaborators about two drugs that have been used “off-label” in to assist the treatment of anorexia nervosa. Fluoxetine, the antidepressant best known as Prozac, and olanzapine, one of the newer “atypical” antipsychotics, are both occasionally used in combination with non-pharmacological treatments, despite scarce scientific evidence of their effectiveness.

“That was the sort of information we as basic scientists weren’t so aware of, because we don’t see patients and there’s no approved treatment,” Dulawa said. “But it’s not really clear which of those drugs are actually helping or not.”

Klenotich pre-treated her mice with either fluoxetine, olanzapine, or a saline control, then exposed them to the ABA conditions for two weeks. The weight of each mouse was tracked on a daily basis, and when it fell to 75 percent of their original weight, they were removed from the study. Those mice who were able to maintain their weight above this drop-out point were considered to have “survived” the anorexia-like conditions. Over the course of many experiments, it became clear that one drug was working far better than the other in improving the survival of the ABA mice.

“We found over and over again that olanzapine was effective in harsher conditions , less harsh conditions, adolescents, adults — it consistently worked,” said Klenotich, a graduate student in the Committee on Neurobiology.

The research also addressed a common concern about olanzapine and other atypical antipsychotics, which are known to carry a side effect of sedation and weight gain. The doses used in the mice were proportionally much lower and shorter in duration than those commonly used to treat diseases such as schizophrenia, and no sedative effects were detected, as measured by how often the mice used their exercise wheels. Demonstrating that this debilitating side effect may not be an issue at the lower levels used for treating anorexia nervosa could reassure a patient population that is already very skeptical and reluctant to take medication, Le Grange said.

“I think the clinical field is certainly very ready for something that is going to make a difference,”said Le Grange, professor of Psychiatry & Behavioral Neuroscience. “I’m not saying there’s a ‘magic pill’ for anorexia nervosa, but we have been lacking any pharmacological agent that clearly contributes to the recovery of our patients. Many parents and many clinicians are looking for that, because it would make our job so much easier if there was something that could turn symptoms around and speed up recovery.”

But olanzapine may just be the starting point. Klenotich and Dulawa now hope to use the ABA model to find out which of the drug’s many neurobiological effects are responsible for its anti-anorexic activity, testing more specific drugs and looking at measures such as gene expression for clues about how it works in the brain. The results could inform the design of an even more effective drug with fewer side effects for patients with anorexia nervosa.

“We can dissect the effect and hopefully identify the mechanisms of action, and identify what receptor systems we want to target,” Klenotich said. “Hopefully, we can develop a newer drug that we can aim towards the eating disorders clinic as an anorexic-specific drug that might be a little more acceptable to patients.”

Developing such a drug might not only directly help clinical treatment of anorexia, it could also change public perception of the eating disorder. Unlike other psychiatric conditions, such as depression or autism, that have recently gained acceptance as diseases of the brain, many people continue to believe that anorexia nervosa and other eating disorders are caused by societal or family influences. By improving our understanding of the neurobiology of the condition — and the pharmacology of how to address it — researchers may change attitudes about the nature of this serious disease.

“Families with anorexic children could become more aware that anorexia is a brain disorder with a large genetic contribution, and doesn’t merely result from their child having psychological problems,” Dulawa said. “Removing that stigma might make parents more willing to advocate for more funding for the disorder, and less likely to blame their child or themselves.”


Klenotich, S., Seiglie, M., McMurray, M., Roitman, J., Le Grange, D., Dugad, P., & Dulawa, S. (2012). Olanzapine, but Not Fluoxetine, Treatment Increases Survival in Activity-Based Anorexia in Mice Neuropsychopharmacology DOI: 10.1038/npp.2012.7

About Rob Mitchum (525 Articles)
Rob Mitchum is communications manager at the Computation Institute, a joint initiative between The University of Chicago and Argonne National Laboratory.
%d bloggers like this: