International Study Provides New Genetic Clue to Anorexia

The largest DNA-sequencing study of anorexia nervosa has linked the eating disorder to variants in a gene coding for an enzyme that regulates cholesterol metabolism. The finding suggests that anorexia could be caused in part by a disruption in the normal processing of cholesterol, which may disrupt mood and eating behavior.

"These findings point in a direction that probably no one would have considered taking before," said Nicholas J. Schork, a professor at The Scripps Research Institute (TSRI). Schork was the senior investigator for the multicenter study, which was published recently online ahead of print in the journal Molecular Psychiatry.

Anorexia affects up to 1 percent of women, and has an estimated mortality rate of 10 or more percent, making it perhaps the deadliest of psychiatric illnesses. Anorexics severely restrict eating and become emaciated, yet see themselves as fat. Individuals with anorexia nervosa tend to be perfectionistic, anxious or depressed, and obsessive, said Walter Kaye, a co-author on the study, professor at the University of California (UC), San Diego School of Medicine and principal investigator of the Price Foundation Genetic Studies of Anorexia Nervosa.

How the disorder develops is still not fully understood. Anorexia predominantly affects girls and young women (the estimated gender ratio is nearly 10:1) and appears to be influenced in part by cultural factors, stress, puberty and social networks. Yet twin studies suggest that genetic factors have the largest influence.

The big mystery has been: what are those genetic factors? Gene-association studies of anorexics have so far produced few replicable findings. Researchers suspect that many genes can contribute to the disorder- and thus only large studies will have the statistical power to detect those individual genetic influences.

For this project- the largest-ever sequencing study of anorexia- Schork worked with an international team of collaborators representing more than two dozen research institutions. The many contributors included first author Ashley Scott-Van Zeeland from The Scripps Translational Science Institute and Scripps Health in La Jolla, California; Kaye and Pei-an Betty Shih from the UC San Diego; Andrew Bergen from SRI International in Menlo Park, California; Wade Berretini from the University of Pennsylvania; and Pierre Magistreti from Ecole Polytechnique Fe'de'rale de Lausanne. The project made use of genetic information from more than 1,200 anorexia patients and nearly 2,000 non-anorexic control subjects.

For an initial "discovery" study in 334 subjects, the researchers catalogued the variants of a large set of genes that had already been linked to feeding behavior or had been flagged in previous anorexia studies. Of more than 150 candidate genes, only a handful showed statistical signs of a linkage with anorexia in this group of subjects.

One of the strongest signs came from the gene EPHX2, which codes for epoxide hydrolase 2- an enzyme known to regulate cholesterol metabolism. "When we saw that, we thought that we might be onto something, because nobody else had reported this gene as having a pronounced role in anorexia," said Schork.