Chapter 1: The Genetic Basis of Bipolar Disorder

(updated 12/2014)


Whereas 10 years ago it was hoped that a single gene for bipolar disorder might be found, it is now clear that many genes are involved: 226 in a recent review.Nurnberger And they do not belong exclusively to bipolar disorder. But some are remarkably well understood.

Link to Chapter 2


Don’t worry, I won’t snow you with too much basic biology. But you do remember, you knew this once: there are 23 human chromosomes; 22 pairs, one of each from mom and dad, plus the X and the Y sex chromosomes — unless you have two X’s and no Y, in which case you have more genetic material overall and therefore more responsibility to save the planet (that’s the female of the species, guys).

At least one psychiatric illness is caused by a single gene: Huntington’s disease. But so far, nothing else in psychiatry has proved to be that simple. And unfortunately, bipolar disorder seems to be an opposite story, in which many genes are involved. Worse yet, it appears that any given individual can have one of many different combinations of these genes, so that there are many different bipolar disorders, quite literally.  With 226 genesNurnberger you could make about 35,000 different combinations…

Bipolar Genes: a table of examples

Some of the main genes associated with bipolar disorder are shown below. (In this table, we are looking at individual genes, not positions on a chromosome. ) Don’t worry, you don’t need to understand any of the details. The point is to show off how well some of these gene differences are now understood.

For example, in the fourth column of the table you see that an exact difference in the DNA sequence has been identified for these particular genes, and the fifth column shows the impact of these variations.

Gene Name Domain Varia­tion Affects:
SERT Serotonin Transporter (featured in Chapter 1 of Depression is not a Moral Weakness) Serotonin re-uptake Gene length dif­ference Depression, anxiety, alcohol use, treatment response
GSK-3B Glycogen Synthase Kinase 3-beta Biological clock, atro­phy / trophic balance Copy number (base pair repeat) Bipolar disorder risk; lithium response
BDNF Brain-Derived Neuro­trophic Factor Atrophy / trophic balance Base-pair vari­ations (single amino acid substi­tutions) Memory, lithium response
Rev-erb A Rev-erb alpha Biological clock Base-pair vari­ations lithium response
CACNA1C calcium channel subunit Amygdala activation Base-pair vari­ation anxiety, paranoia in bipolar disorder
ANK3 Ankyrin 3 “scaffold protein”, localized ion channels Base-pair variation Anxiety, novelty seeking


“Bipolar” genes overlap other conditions

The gene diagram below illustrates this theme well. In the left column are represented genes known to be associated with bipolar disorder. Actually, these are not individual genes, but rather positions on the various chromosomes. At these positions are found particular gene sequences which appear to differ in people with bipolar disorder. In the right column are chromosome positions known to be associated with schizophrenia. In the middle column are chromosome positions in which particular genetic sequences are associated with symptoms shared by both conditions, such as delusions, hallucinations, and abnormal thought processes (as you probably know from reading elsewhere on this website, these are symptoms of Bipolar I, not Bipolar II. The latter shares some of the genes of Bipolar I, but clearly not all of them, because it does not share these psychosis genes at all).

genes1 (1)

As you can see, any given individual (shown here by the black ellipticals circles) could have one of many different combinations of genes. You would expect that different combinations would produce different manifestations.

You can see in this diagram that some of those variations share genes with schizophrenia. For example, someone who had several genes from the left column, but one or two from the middle column, might have symptoms that look a bit more like someone with schizophrenia than someone whose bipolar disorder was associated with genes from the left-hand column only. in other words, the genes in the left column represent relatively “pure” bipolar disorder.

Imagine this kind of diagram with 266 genes in it. Imagine all the variations you could find: it would be hard to find any two people who were the same, in fact. More like snowflakes than species of birds where all the members look so alike (I guess maybe the birds don’t see it that way; they probably think we humans all look alike).

I hope this helps you understand that “Bipolar I” is thus just a really extreme version of all the bipolar variations: it’s so extreme that it looks almost the same every time it shows up in an Emergency Department. So it can be labeled “Bipolar I” and doctors can agree on the diagnosis. But all those other variations that aren’t so extreme, that look much less like Bipolar I? Well, you see what I’m driving at: it’s a wonder we can have any sort of agreement at all about what’s “bipolar” and what’s not.

Okay, that’s it for basics of bipolar genetics.  Bail out now!  to Chapter 2 on Brain Differences in Bipolar Disorder  – – unless you’re really interested in bipolar genetics, in which case, read on…

Not “bad genes” but “plasticity genes”

The search for genes associated with bipolar disorder is further complicated by the overlap between genes that confer bipolar risk and genes that confer “plasticity”.

Plastic means changeable or shapeable.  Plasticity genes allow individuals to respond more directly to environmental experience, to mold themselves to their environment and potential future environments based on past experience. These include the serotonin transporter gene (SERT), the gene for brain-derived neurotrophic factor (BDNF), and others.

The serotonin transporter (SERT) gene length difference has been extensively investigated in relation to mood and anxiety disorders. You’ll find a full story about the SERT gene in the section entitled Depression is not a Moral Weakness. Reviewing briefly: the short version of the SERT gene is associated with an increased risk of depression in the face of life stresses, but only in the context of adverse childhood experiences. Benign childhoods appear to completely mask the gene length difference effects.Caspi These findings were recently replicated in bipolar disorder.Benedetti

Similarly, a substantial literature associates the gene for brain-derived neurotrophic factor (BDNF) with mood disorders and bipolar disorder in particular. Oh, sorry, you haven’t run into BDNF yet?  Go learn about it then return here (I’ll open that Introduction to BDNF in another window for you).

A base pair difference in the gene (single nucleotide polymorphism, or SNP) leads to insertion of a methionine in the BDNF protein instead of a valine.  The methionine variant is associated with increased susceptibility to Alzheimer’s, Parkinson’s, depression, eating disorders – and bipolar disorders.Bath  In bipolar disorder, carriers of the methionine-yielding allele have significantly higher suicide attempt rates. Kim

But although these alleles confer risk of a potentially lethal disorder, they must confer some benefit, else they would have been selected out evolutionarily long ago (given that they act in young and middle age).  Indeed,  these are not “bad genes, not even “susceptibility genes”. In some contexts they are beneficial, or protective.

For example: inheriting the Met allele of the BDNF gene looks like a bum deal, right?  But two studies have found that in the context of family maltreatment, inheriting the BDNF Met allele lowered susceptibility to adult depression — in individuals who carried two short versions of the SERT gene.Grabe, Comasco  Similarly, the short version of the SERT gene appears to confer a degree of vigilance and capacity to handle rapid changes in stress that is evolutionarily advantageous.Homberg, Dobson

Given that there are multiple “plasticity” genes (at least 4 beyond SERT and BDNF), the interactions are sure to be extremely complex. Nevertheless, one thing is clear: these genes interact with childhood environment to affect risk of developing mood disorders.

Epigenetic modification of bipolar genes

Never heard of or don’t understand epigenetic modification? Better read about that, it’s an amazing story.

But as of 12/2014 there is almost no data on epigenetic modifiation of bipolar-related genes.  Only a few have been tentatively identified.KatoThe epigenetic modification story is still mostly about depression and stress-related genes. Sure, these matter in bipolar disorder as well.

By contrast, the next and last section is about a gene where more detail is understood.

GSK-3b: an brief example of how much is known

Glycogen synthase kinase 3-beta is an enzyme which appears repeatedly at the crossroads between pathways associated with mood problems. Exactly how it works in creating mood symptoms is not yet known. But many of the known treatments for mood disorders work through pathways that pass through this enzymatic step, as shown in the following diagram. This diagram is explained in Chapter 10 of the Depression essays.

(note the pink GSK-3 rectangle in the lower left-hand corner):

10AllP16 (1)

If you continue through this story about what causes bipolar disorder, you are going to run into this enzyme again in the section on the biological clock. Recently it was discovered that lithium works by inhibiting GSK3-beta and thereby restoring normal cycling of the biological clock. (Lithium works in other ways as well, but this may be one of the most important).

Link to Chapter 2: Brain differences