In Search of the Elusive Bipolar Genes: A New Study Sheds Interesting Light

John McManamy Health Guide
  • Will scientists ever crack the bipolar gene riddle? Back in 1987, they thought they had. According to Nature, the august journal that published Watson and Crick’s ground-breaking research, bipolar “can be caused by a single gene.” 


    Not only that, researchers localized “a dominant gene conferring a strong predisposition to manic depressive disease to the tip of the short arm of chromosome 11.”


    Subsequent studies failed to replicate the findings. In retrospect, we can only marvel at the idea of very smart people - less than three decades ago, at that - actually believing that an illness as complex as bipolar could be caused by one rogue gene. 

    Add This Infographic to Your Website or Blog With This Code:


    Over the years, the gene quest grew ever more sophisticated, but yielded signals very difficult to separate from the background noise. These included linkages to a variety of “psychosis” genes that also turned up in the genomes of those with schizophrenia.


    This was the early 2000s. The researchers I talked to all told me that bipolar was “multifactoral” and “heterogenous.” In other words, bipolar had many causes and that more than one gene would be involved.


    My impression was they were expecting to find somewhere in the neighborhood of 20 gene variations and that someone with bipolar might have five or six or seven of them. My collection of bipolar genes could very well be completely different than your collection of bipolar genes.


    There were also the twin issues of “penetrance” and “expression.” You could have a bad variation of a certain gene, but not necessarily the illness. This is easy to figure out with a “one gene to one illness” correspondence, such as cystic fibrosis. Bipolar is a lot more problematic.


    Despite the hurdles involved, I felt a sense of optimism from the research camp. The human genome had been sequenced, and gene chip technology was accelerating at an exponential rate. The researchers were hot on the scent.


    In hindsight, we can see a major flaw in their reasoning. This had to do with their naive attempt to link “gene” to “phenotype.” Phenotype refers to observable characteristics or traits. In the context of bipolar, this equates to classic mania and depression and psychosis. The catch, as I heard a number of speakers at conferences remark, is that our genes do not encode with the DSM in mind.


    At the time, an eye-popping 2003 study was fresh on everyone’s minds. I have reported on this study many times here and elsewhere. Rather than look for, say, a “depression” gene, the researchers fished further upstream. They found a gene implicated in how the brain reacts to environmental stress. Lo and behold, one of the behavioral results was depression.


    As opposed to “phenotype,” the researchers were looking to “endophenotype.” In the context of bipolar, these are more common gene variations - such as genes that regulate sleep or stress or cognitive processing - that may result in classic bipolar symptoms further downstream.


    Add This Infographic to Your Website or Blog With This Code:

    These endophenotypes may seemingly have nothing to do with mental illness. Schizophrenia researchers, for instance, have been hot on the trail of nicotine addiction and quirky eye movements.


    Technically endophentypes are phenotypes in their own right. So maybe we need to completely deconstruct bipolar. Break it down into as many underlying phenotypes as we can, find a population or two to study, measure everything, perform brain scans and gene scans, statistically analyze the crap out of everything, and see what happens.


    Fast forward to 2009. I was in Pittsburgh at the International Conference on Bipolar Disorders. There, I ran into Carrie Bearden of UCLA. I had corresponded with her in the past. I had also run into her at another bipolar conference and at a schizophrenia conference.


    I asked her what she was up to. She told me, in effect, she was commuting to Colombia, way out in the middle of nowhere, studying an isolated population in a gene study she was involved in.


    Cool, I thought. It is common practice for gene researchers to seek out isolated populations - Icelanders, the Amish, you name it - so I thought no more of it. Really, I need to learn the art of asking follow-up questions. 


    The scene is set ...  


    Four days ago, Dr Bearden and her colleagues published the first of their results in JAMA Psychiatry. The researchers investigated isolated populations in Costa Rica and Colombia. This involved 181 individuals with bipolar, plus their non-bipolar relatives. Rather than look for links to the overall bipolar syndrome, the researchers teased out 169 component phenotypes.


    For example, under the subdomain of “temperament,” the researchers measured “delusion proneness,” “perceptual creativity,” “affective temperament,” and “impulsivity/risk-taking.” 


    “Affective temperament” broke down into five phenotypes, including anxiety, cyclothymia, depressive, hyperthymia, and irritability. 


    “Impulsivity/risk-taking” involved aggression, sensation-seeking, and other behaviors. 


    Under “neurocognition,” the researchers looked at a host of phenotypes involving long-term memory, executive function, and working memory. 


    The researchers also checked out sleep activity and circadian rhythms, which they will report on later. 


    Now for the results: Three-quarters of the 169  phenotypes that the researchers investigated turned out to be “significantly heritable” and nearly a third bore an association with bipolar I. A quarter were both heritable and associated with bipolar I.


    The researchers also performed brain scans on the subjects, which turned up a number of anomalies, including reductions in cortical thickness in various regions of the prefrontal and temporal cortices.


    The next step is to begin identifying specific genes based on genomic data collected from the study subjects and map and cross-map these to the brain and phenotype hot spots.


    Will we find some convincing suspect genes? Who knows? Are we bound to learn something? Most definitely.

Published On: February 16, 2014