Depressing conclusion as new study reverses 25 years of research

New techniques for identifying the genetic roots of disease have led researchers to conclude that 25 years worth of research studying a set of genes thought to be strongly associated with depression is wrong, according to a recent study in the American Journal of Psychiatry.

Countless hours, and an estimated $250 million US, have been spent over the past 25 years investigating a connection between certain genes and mental illness. But that connection, new research suggests, simply isn't there.

"The fundamental problem was this sort of misplaced optimism — that was very defensible at the time — that we'd find genes that had big effects,"  Richard Border, the study leader, told Quirks & Quarks host Bob McDonald.  Border is a PhD candidate in statistical genetics at the University of Colorado Boulder.

A quarter century ago, researchers identified genes to investigate that were related to systems they thought were important to depression, such as the serotonin system, which has a powerful effect on mood. Studies of these "candidate genes" looking at hundreds of people with and without depression, seemed to show a correlation between certain gene variants and major depressive disorder.

The bulk of these studies took place when technology and cost made it possible to study individual genes, or certain areas of the genome, but not the genome in its entirety.

What we found is that there's very little support at all for these historic hypotheses.- Richard Border, University of Colorado Boulder

"The most studied candidate gene with respect to depression, by far, is the serotonin transporter gene. And in the 90s and early 2000s, it made a lot of sense to look at this variant," said Border.

The first generation of studies looking for connections between variants of this gene and depression seemed to show great promise. But in followup work scientists had difficulty replicating the results. Border said that some researchers, persuaded that there had to be some kind of connection, would look for reasons to explain away the problems in replication — making their hypothesis and explanations more and more convoluted and complicated.

"In the meantime, in the last 10 years there's been this sort of genome-wide revolution in genetics where it's now very cost effective to have large samples where you can look across the entire genome at millions of variants in very large samples," added Border.

Investigating 'candidate genes' with today's genetic tools

Border used this new capacity to perform what is known as a genome wide association study, or GWAS. He looked at 18 of the "candidate genes" that early studies had connected with depression, which had been thoroughly studied ever since. 

They re-examined the associations between these 18 genes and depression in very large databases of hundreds of thousands of people.

"What we found is that there's very little support at all for these historic hypotheses," said Border.

We have some colleagues in the broader behavioural sciences ... who are unfamiliar with this great deal of scepticism in mainstream genetics about this body of research.- Richard Border, University of Colorado Boulder

Border says he — and many of his colleagues familiar with more recent findings in genetics, weren't really surprised by this finding. In the genetics community there's increasing understanding that complex traits like depression rarely depend on the strong effects of just a few genes. 

"What's interesting," added Border, "is we have some colleagues in the broader behavioural sciences, who are maybe even doing candidate gene research still to this day themselves, who are unfamiliar with this great deal of scepticism in mainstream genetics about this body of research."

In fact, Border says, it was bridging this gap that was part of his motivation to do this study, "to definitely address the issue of candidate genes."

Genetics of Depression 2.0

One of the things this study did was underline the new consensus about the genetic complexity of depression. 

According to scientists at the forefront of the research, there are hundreds, maybe even thousands of genes contributing to depression — each one playing a very small part.

"It's estimated that about a third of the risk of depression is caused by genetic factors," said Andrew McIntosh, a professor of psychiatry from the University of Edinburgh. "The remainder is due to a number of other factors — some of it is chance and the rest of it is probably something to do with the environment."

So far, McIntosh and his colleagues have discovered 102 genetic changes and 269 genes associated with depression, none of which are the historical candidate genes that had been previously studied, and all of which would contribute in only very small ways.

They still expect to discover many more genes associated with depression to bridge the gap between what they've found and what they expect to find based on estimates of the genetic contribution to depression.

New genetics, new hope

These contributions can be subtle, but should still ultimately be useful in understanding depression.  For example, by combing through findings from GWAS studies, McIntosh said they've discovered genes involved in cellular processes that maintain the communication from one nerve cell to another via the neurotransmitter glutamate. This could be an important pathway for understanding depression. 

McIntosh said there are a number of similar emerging findings that could be new targets for developing therapies. 

They also discovered a number of environmental risk factors that interact with genes that came up in their GWAS study.

"It looks from our study like being overweight ... makes you more likely to develop depression," said McIntosh.

They also found that a tendency to focus on more negative emotions also adds to the risk of developing clinical depression.

"We're reasonably confident that as sample sizes develop and grow, we'll be able to identify other risk factors — other environmental things — that are potentially modifiable," said McIntosh.