[size=9px]Sep 29th 2005
From The Economist print edition[/size]
Anorexia and bulimia may be autoimmune diseases?and so may several other psychiatric illnesses
SOMETIMES, the immune system works in mysterious ways. During an infection one of its roles is to produce antibodies designed to attack and eliminate the invading bugs. However, in certain unlucky individuals the body also develops so-called autoantibodies which attack its own tissue, sometimes with devastating effects. The result is known as an autoimmune disease, two well-known examples of which are type-1 diabetes and multiple sclerosis. But there is a widespread suspicion among researchers in the field that a lot more diseases than these have an autoimmune component. In particular, they think, a number of illnesses usually labelled as ?psychiatric? are actually, at bottom, the result of autoimmunity.
Until now, that suspicion has been based on correlations between certain sorts of infection and certain sets of psychiatric symptoms. But work just published in the Proceedings of the National Academy of Sciences by Serguei Fetissov of the Karolinska Institute in Stockholm and his colleagues has tied the connection more tightly for two psychiatric eating disorders?anorexia nervosa and bulimia nervosa.
Dr Fetissov's work suggests that abnormal levels of autoantibodies against hormones called melanocortins are a crucial part of the cause of these two diseases. Melanocortins are small protein molecules that carry messages between nerve cells in the brain. They are involved in regulating a variety of complex behaviours, including social interactions, stress responses and?most importantly in this context?food intake. So it is easy to see how interfering with them could cause anorexia and bulimia.
Shooting the messenger
To test this idea, Dr Fetissov and his colleagues analysed blood serum from three groups of women (both anorexia and bulimia are more common in women than in men). One of these groups consisted of people diagnosed as anorexic. The second was composed of individuals diagnosed as bulimic. The third contained people with no eating disorder.
The researchers looked to see whether there was any relationship between the levels of autoantibodies to melanocortins in these women and their expression of particular psychological traits?such as ?Drive for thinness?, ?Body dissatisfaction? and ?Perfectionism??which are associated with eating disorders and which can be measured using a specially designed scoring system.
What they found was intriguing. There was not one relationship, but two. The level of autoantibodies to melanocortins was positively correlated with anorexia, but it was inversely correlated with bulimia. These opposite correlations make sense. Although both disorders are associated with depression and self-doubt, anorexia involves a constant refusal to eat, whereas bulimia is a ?diet-binge-purge disorder? that includes periods of excessive consumption. The molecular triggers of the two could thus easily be opposites.
The ultimate cause of the altered levels of autoantibody in anorexics and bulimics is unresolved as yet. However, according to the researchers, a clue may lie in the fact that micro-organisms, too, work in mysterious ways. In the world of bacteria and viruses, a strategy called molecular mimicry is common. In this, pathogens evolve to produce pieces of protein similar to those of their hosts, as a way of confusing that host's immune system. But the immune system is not always fooled, and in making antibodies to the ?camouflage? proteins it sometimes turns out weapons that also attack the useful proteins that are being mimicked.
Two common gut bacteria, Escherichia coli and Helicobacter pylori, and also the influenza-A virus, are particularly adept at playing the evasive game of molecular mimicry, and the team is now looking at possible connections between different gut bacteria and autoantibodies against melanocortins to see if they can pin down which, if any, of these bugs might be responsible.
That is not to say, even if Dr Fetissov's idea is correct, that autoimmunity is the whole story. Both anorexia and bulimia are known to go hand in glove with particular personality characteristics which are not directly related to the disease. In anorexics, striving for perfection and conscientiousness are common non-pathological traits. In bulimics, such traits include risk-taking behaviour and problems with impulse control. So there appear to be predisposing factors at work, as well as the triggering effect of the autoantibodies.
Nevertheless, given the range of behaviours regulated by melanocortins and other, similar, messenger molecules, the suspicions that other psychiatric disorders?in particular, obsessive-compulsive disorder?are partly or wholly the product of a similar process seem entirely plausible. Dr Fetissov's work also adds weight to the idea that two other neurological diseases, schizophrenia and Tourette's syndrome, have an autoimmune component. In the case of these diseases, the damage seems to be caused irreversibly in the womb, suggesting that any autoantibodies involved are attacking structural molecules rather than messengers (attacks on structural molecules are the cause of multiple sclerosis, though they involve a different part of the immune system). That gives little hope for treatment. But in the case of anorexia, bulimia and, possibly, obsessive-compulsive disorder, Dr Fetissov's work opens a new line of thinking about how these diseases might be treated.