Genetic Risk Factors Found For Major Mental Disorders — But We All Have Them

A paper published February 28 in the Lancet reported the identification of genetic variations (SNPs) that mark an increased risk for ADHD, autism spectrum disorder, bipolar disorder, major depressive disorder, and schizophrenia. The paper was based on meta-analysis of four previous genome-wide association studies that had targeted these conditions individually. From the whole-genome data, the group identified four specific variations that affect the overall risk of all five disorders, at least in those of European ancestry.

Two of these variations corresponded to the genes CACNA1C and CACNB2, which work together to allow calcium into a cell in response to changes in the tiny voltage across the cell's membrane. These genes may control nerves in the brain, but they also control blood pressure. Blood pressure has been identified one of several cardiometabolic factors linked to the risk of schizophrenia. The authors have suggested that it may be worth studying existing high blood pressure medications for psychiatric use.

Two other genes are more difficult to identify. This is because of linkage disequilibrium, a common effect in the human genome which makes it hard to tell which of a set of adjacent variations is responsible for psychiatric trouble because they all match. Ordinarily, SNPs are shuffled by the swapping of genes between related chromosomes. In linkage disequilibrium, that doesn't happen, so looking at any SNP in a broad region of DNA — in one of these examples, a million "letters" of the DNA code — gives the same statistical result. As a result, the precise genes involved in those SNPs cannot be identified, but the possibilities are narrowed down from 30,000 genes to 30.

Unfortunately, the power of whole-genome studies is limited by the extraordinary degree of certainty needed to obtain usable results.  Because the Consortium tested gene sequences at 1.25 million SNPs, for them to prove an association they needed to use a threshold of about 1 in 24 million, allowing the traditional (though arbitrary) 5% chance that one of the genes they report is merely a chance association. This is achieved only by examining all these sequences from 60,000 people, and even that gives only enough statistical power to find four genes.

However, the Consortium was able to delve deeper into the statistics to analyze a restricted subset of genes that had previously been associated with one of the five disorders, using that data to deduce that one of them, NT5C2, is linked with all five conditions. This genetic variant, previously identified as a risk factor for schizophrenia, helps to regulate purine metabolism, a process linked to several conditions such as purine autism. They evaluated eleven others, finding that some identified with bipolar disorder or schizophrenia appear specific to those conditions, while several others linked with schizophrenia have a broader impact on autism or major depressive disorder.

The reason for choosing these five conditions was only that they were common and large amounts of data had been collected. It is possible to argue, as many have done, that there may be overlap or misdiagnosis linking the conditions, but it is hard to describe them all as forms of one genetic syndrome. It was even proposed in 2009 that autism and schizophrenia were opposite ends of a mental spectrum, based on overall variations in gene copy number seen in patients with the two conditions. Maybe what these researchers have really done is to identify the relatively rare non-risky versions of these genes as contributors to some physical mechanism of psychological normalcy.

It is incorrect to brand these variants as "genetic glitches" or "faulty genes." These are actually very common variants — all but one of the five genes mentioned above are the most common form of the gene found in European populations, and even that one occurs one-third of the time. Because most people have two copies of each gene, there is less than a 2% chance that we carry them all — but less than a one in a million chance we don't carry any of these risk factors.

Genetic variants this common are the winners of the cruel game of natural selection, so they cannot be seen as pathological. However, genetics is like a pick-up ball game: a player who passes the ball frequently may be an asset in most games, but not if the teammates who turn up that day can't catch. Most likely these genetic variants make up for a slightly increased risk of mental illness with some equally small benefit. However, we should also consider the uncertain but fascinating possibility that the two regions showing linkage disequilibrium have been the target of recent selection, and that we might actually be witnessing one tiny aspect of the ongoing evolution of the human mind.