This is part II in my series of posts on the role of dopamine in anorexia nervosa. (You can find the first part, which covers the basics of dopamine signalling, here.) In this post I’m going to discuss the findings from preclinical studies (studies in animal models).
I don’t think I’ve talked about animal models of anorexia nervosa before on the blog, but believe or not, they exist. The most well-known one is called activity-based anorexia (ABA). ABA works like this: rats are simultaneously restricted in the amount of food they can eat and given access to a running wheel. As the rats experience a reduction in their caloric intake, they begin to spend more and more time running on the wheel. A similar model with basically the same premise is called starvation-induced hyperactivity. These models are thought to mimic both the restriction/weight-loss and excessive exercise components of anorexia nervosa.
The ABA model has been used to study various aspects of anorexia nervosa, including the relationship between AN and over-exercise. Given that dopamine is known to be involved in feeding behaviours, reward, motivation, and motor activity, its role in anorexia has also been investigated using the ABA model.
First, a quick reminder of the two theoretical models that have been proposed for the role of dopamine in anorexia nervosa:
There are at least two main [competing] theoretical models explaining the role of dopamine in AN. The first suggests that food restriction in AN is rewarding. [The second model] proposes that the reduction in food consumption observed in AN is because of a lack of pleasure associated with food intake… These two models fail to recognize the different aspects of reward and motivation and may be too simplistic. However, they could be a good starting point for a more detailed study of the role of dopamine in AN.
So, what have researchers studying rats and mice found?
An early study in 1990 by Broocks et al. found that a reduction in feeding (leading to a 30% drop in body weight) led to decreased dopamine turnover in a part of the rat’s hypothalamus. Dopamine turnover is a way to measure the activity of dopaminergic neurons, whereby increased dopaminergic activity would lead to increased dopamine turnover, and vice-versa. Interestingly, they found that hyperactivity (increased running on the wheel) led to an increase of dopamine in the hypothalamus.
The authors speculated that “running is accompanied by some kind of pleasurable, rewarding effect, or–the other way round–that it might compensate for the consequences of an unpleasurable condition as the state of hunger certainly is.”
Another study, in 2008, found that between two different inbred groups of mice, one was more susceptible to develop ABA, and this susceptibility seemed to be modulated by dopaminergic signalling. The mice that were more susceptible to develop ABA showed an increase in D2 receptor expression in a particular region of the basal ganglia (caudate-putamen).
These findings were concordant with the second theoretical model of etiology of AN. The authors concluded that in response to restricted feeding, mice [that are more susceptible to ABA] might have engaged in a compensatory behavior in an attempt to manage the disturbed state characterized by the lack of pleasure originating from food.
These studies suggest that if increased dopamine receptor expression or dopamine signalling leads to the development (or higher susceptibility of) ABA (and thus anorexia nervosa in humans), using something to block dopamine signalling would prevent or ameliorate ABA.
Supporting this idea, in 2009, Verhagen and colleagues used a drug to block dopamine activity (though the drug blocks other receptors, too) and found that it led to a decrease in wheel running activity, increased body weight and food intake in ABA animals. A similar finding was reported by Adams et al (2009), who found that dopamine antagonist chlorpromazine prevented wheel-induced suppression of eating. Finally, Hillebrand and colleagues (2005) found that olanzapine (Zyprexa) reduced the development of ABA in rats. Interesting, right?
(As a side-note, there have been many studies now looking at whether olanzapine is effective in treating anorexia nervosa. A colleague of mine in graduate school has been involved in at least one of these studies, I may pick her brain about it at some future point and write about it. Though based on this review, antipsychotics, including olanzapine, don’t look promising.)
But hold on.
Although dopamine antagonism has mainly yielded positive results in animal models of AN, other findings showed that anorexia-like symptoms could be improved by an increased, rather than a decreased dopamine turnover.
In one study, injecting a precursor to dopamine (which led to an increase in dopamine) enhanced the rat’s performance on a learning and memory task (Avraham et al., 1996). In another study, injecting the dopamine precursor similarly enhanced the rat’s cognitive abilities, though it did not affect body weight (Hao et al., 2001).
However, this last study by Hao et al did not use the ABA model, instead the rats experienced “self-induced weight loss caused by separation [from their mother] stress.” Perhaps these contradictory findings are due to the fact that in these two studies, the rats did not experience weight-loss as a result of hyperactivity? Indeed, hyperactivity wasn’t really a component of their models. This is simply my guess, I may easily be wrong there.
To conclude, the preclinical studies on the role of dopamine in anorexia nervosa are somewhat contradictory. It seems that more studies support that AN is associated with (note: not caused by) increased dopaminergic activity and that administrating a dopamine blocker reduces the AN-like state. However, some studies have found the opposite: decreased dopamine activity in AN-like states, which is ameliorated by increasing dopamine.
In the next post, I’ll begin to discuss clinical studies which look at the role of dopamine in anorexia nervosa patients.
Kontis, D., & Theochari, E. (2012). Dopamine in anorexia nervosa Behavioural Pharmacology, 23 (5 and 6), 496-515 DOI: 10.1097/FBP.0b013e328357e115