Genetic Study Suggests Anorexia May Have Metabolic Origins

The study shows that individuals with genetics that predispose to anorexia are similar to individuals with genetics that predispose to “constitutional thinness”.

This week, I want to talk about this article, appearing in Nature Genetics – a genome-wide association study of anorexia nervosa which may fundamentally reshape how we think of the disease.

But let me start with how genome-wide association studies, or GWAS works.

Imagine the entire human genome- 23 pairs of chromosomes, all 3 billion base pairs, stretched out linearly, like a road extending through a vast desert.

Prices of Manhattan plots continue to skyrocket. Experts suggest Westchester plots if costs is an issue.

Prices of Manhattan plots continue to skyrocket. Experts suggest Westchester plots if costs is an issue.

Along that road there are around 25,000 street lights – genes, some all bunched up like you might see around a city, and some all by themselves in long, empty stretches. For a given disease or condition different street lights shine brighter or less bright.  But we are flying far above, too far to pinpoint exactly where the street lights are, but we can get a general sense that some parts of the highway are brighter than others.  This is how GWAS works. Instead of sequencing the entire genome of an individual – every inch of that highway, it looks at a number – typically a million or so – representative points (called single-nucleotide polymorphisms or SNPs) and asks if they are a bit brighter than the background.  If so it suggests that a bright streetlight – a causative gene – may be nearby. 

Practically, researchers create cohorts of tens and hundreds of thousands of individuals with a disease or condition and compare the pattern of SNPs in their genomes with those in a control population. What you get is a plot like this – a Manhattan plot.

Think of the high points as bright spots in the genome. Somewhere around those areas, the thinking goes, are genes that affect the condition being studied.

What researchers, led by Cynthia Bulik at the University of North Carolina found, as you can see from the Manhattan plot are 8 “bright spots” that associate with anorexia.  The brightest of all is in an area of chromosome 3 that has 111 genes, highlighting one of the main limitations of GWAS studies. We often can’t tell which genes are really driving the findings.

Take-home number one: There are genetic underpinnings to anorexia.

But we can do more than look at single hot spots, we can look at the entire genetic pattern seen in individuals with anorexia and compare that pattern to GWAS’ that have been done in other conditions.

That analysis revealed some truly surprising results.

Genetic correlations here, not observed correlations (though there are observed correlations). My brain hurts.

Genetic correlations here, not observed correlations (though there are observed correlations). My brain hurts.

First of all, there was very strong similarity between the gene patterns of anorexia and those of obsessive-compulsive disorder.  In fact, no two psychiatric diagnoses are as strongly linked, genetically, as OCD and anorexia.

But the data suggests that some of the underpinnings of anorexia aren’t psychiatric at all. The gene pattern among the individuals with anorexia was negatively correlated with gene patterns that lead to higher body fat and fasting insulin levels.

Let me be really clear here – this isn’t saying that there is an association between Anorexia and low BMI – I mean, of course there is. This is GENETIC information. This says that individuals who are born predisposed to anorexia are ALSO born predisposed to being constitutionally thin. This runs counter to the prevailing wisdom of the etiology of the disease. I’ll let lead author Dr. Cynthia Bulik explain. 

“So my entire career I have been wrong. I always thought that constitutional thinness was protective against anorexia nervosa. That does not appear to be the case because some of the same genetic patterns that might increase the likelihood of you being constitutionally thing might also increase the likelihood of you becoming anorexic.”

This is a paradigm shift. It shows us not only that anorexia has genetic underpinnings.  It also shows us that anorexia may begin FIRST as a disorder of rapid metabolism. This emerging understanding may help address the stigma associated with the disease. Here’s Dr. Bulik Again:

“This is actually giving an explanatory framework to patients and family members to understand why this disorder is A) so deadly and B) so difficult to recover from.”

As the genetics of anorexia become more clear, the hope is that new therapies, perhaps targeting the non-psychiatric components, will emerge.

 This commentary first appeared on