A powerful piece of medical dogma is the notion that depression has genetic roots. This is the idea that if your mother suffered from depression, then the chances are that you will too. A closedly associated notion is the “chemical imbalance” in the brain idea. This is the claim that depressed patients do not have enough serotonin. Hence, many antidepressants boost the level of circulating serotonin.
The “depression is genetic” notion holds that there are specific genes that cause the serotonin imbalance. Remarkably, it turns out that depression is not strongly influenced by one’s genes, contrary to years of medical ‘wisdom.’ This iconoclastic conclusion comes from an exceptionally powerful study into the genetics of depression. For decades, scientists have been trying to identify the responsible “candidate” genes. The small set of genes that they believe render carriers highly susceptible to developing depression. Turns out we were wrong!
One gene in particular, the serotonin transporter gene SLC6A4, got a lot of attention. In 2003, Caspi et al. published a now famous paper. They summarised their findings, remarking:
Individuals with one or two copies of the short allele of the 5-HT T promoter polymorphism exhibited more depressive symptoms, diagnosable depression, and suicidality in relation to stressful life events than individuals homozygous for the long allele. This epidemiological study thus provides evidence of a gene-by-environment interaction, in which an individual’s response to environmental insults is moderated by his or her genetic makeup.
Caspi et al, 2003
The Caspi research struck a chord and resonated with scientists. It seemed to make good sense, after all. The Caspi paper has been cited 8527 times, indicating its widespread acceptance. However, Caspi and colleagues were wrong, as were hundreds of other researchers who published thousands of papers in the belief that they had found evidence to show a linkage between genetics and depressed mental space. Richard Border, Matthew Keller and colleagues, all from University of Colorado Boulder, turned that idea on its head. They effectively debunked the notion that the candidate genes identified thus far play a major role in depression.
How did they reach this remarkable conclusion? Border et al. used multiple large samples ranging in size from 62,138 to 443,264 people, in total drawing data from 621,214 individuals. That’s very impressive! Sample sizes of that magnitude have the statistical power to answer the questions about genetics and depressed mood and avoid false positive results (type I errors). By comparison, Caspi et al. had a sample size of 847 subjects.
Border et al. measured depression in many different ways to ensure that their study was not limited by narrow definitions of depression. They considered diagnosis, current severity, lifetime symptom count, duration of depression, the presence of recurrent symptoms, and so on. To be certain that they identified all and any associations between depression and genetics, they also used very liberal criteria for statistical significance.
They took the 18 genes most commonly suspected to cause depression. They then examined the interaction between those genes and their multiple definitions of depression, but they found nothing significant. In fact, they found that the association between depression and suspect candidate genes was the same as the association between the condition and randomly selected genes. Basically, this means that the previously reported associations were chance findings.
As common sense would predict, they did find an association between exposure to psychological trauma and depression, but they did not find any gene-environment interaction between candidate genes and psychological trauma. Nor did they find any association with common single nucleotide polymorphisms across each of the candidate (suspect) genes.
Border and colleagues highlight the inadequacies of previous research. They demonstrate that the previous findings were, in fact, false positives. They conclude:
“that it is time for depression research to abandon historical candidate gene and candidate gene-by-environment interaction hypotheses.”
The findings of Richard Border and colleagues highlights the danger of a knee-jerk approach to scientific reductionism in combination with small sample sizes. As anyone who has suffered from depression knows, depression is a highly complex emotional space and is influenced by a multitude of factors, internal and external. Senior author, Prof Matthew Keller, an associate professor of psychology and neuroscience, summed it up nicely by saying,
“We are not saying that depression is not heritable at all. It is. What we are saying is that depression is influenced by many, many variants, and individually each of those has a minuscule effect.”
Prof M Keller
Personally, I think that these are liberating findings. Many depressed patients and their treating health care professionals have been caught in the mental trap of believing that the patient’s low mood was strongly driven by genetic factors. On that view, patients were doomed to a lifetime of depressed mood by their DNA and only antidepressant medication could help to ease the pain. Border’s findings position the cause of depressed mood where it belongs; in the mind.
That underscores the importance of treating depression with psychotherapy. That is not to say that it is easy to treat depression, but it is easier to shift entrenched mental habits than it is to change DNA (with apologies to all the CRISPR-Cas9 editing fans out there).
