Living the Scientific Life (Scientist, Interrupted)

Some of you, like me, suffer from bipolar disorder or might know someone who does, so I thought I’d take this opportunity to write a little about the creation of a mouse model to study the genetics that are thought to underlie the manic phase of bipolar disorder — a phase that has not been well understood so far.

Bipolar disorder, or manic-depressive illness, is a psychiatric condition that affects a person’s moods. Typically, a person who suffers from a classical bipolar disorder (Also known as bipolar affective disorder, type 1) will have periods of depression that alternate with mania or hypomania. Mania and hypomania are characterized by elation, irritability, reduced need for sleep and increased goal-directed behaviors. These characteristic mood swings can be associated with the season; depression being more common in winter with either mania or hypomania typically occurring in summer or at the change of the seasons.

Bipolar disorder in its various forms affects slightly less than 3 percent of the population. It is associated with an increased risk of suicide, substance abuse, and vocational disability. Unfortunately, there are no animal models for bipolar disorder, so it has not been possible to carefully study the genetics of the disorder (there are mice who exhibit depression and thereby act as molecular models for that illness). A paper that was recently sent to me by a reader examines the underlying genetics of bipolar disorder and uses that information to create a mouse model of the manic phase of bipolar disorder so scientists and psychiatrists can study this aspect of the condition.

Image source: doi:10.1073/pnas.0701491104. [Bigger image]

Because of its cyclical nature, bipolar disorder has been thought to be tied to the circadian rhythm. In the suprachiasmatic nucleus of the brain, there are several gene products that dictate one’s circadian rhythm; CLOCK and BMAL1, Period (Per) and Cryptochrome (Cry), and a host of other proteins (see figure 1, above). But which gene(s) are essential to mania?

Recent studies of glycogen synthase kinase-3 beta (GSK3beta) have identified this enzyme as the central regulator of the circadian clock and further, GSK3beta is a known target of the mood-stabilizing drug, lithium [Gould & Manji (2005) Neuropsychopharmacology 30:1223-1237] that is commonly used to treat bipolar disorder. Lithium is known to increase the length of the circadian period in fruit flies, Drosophila, and in rodents. Additionally, valproate (depakote) acts on on the circadian cycle by altering the expression of several circadian genes in the amygdala [Ogden et al., (2004) Mol Psychiatry 9:1007-1029] and treatment with fluoxetine (prozac) has been shown to target this pathway by increasing the expression of Clock and Bmal1 genes in the hippocampus [Manev & Uz (2006) Trends Pharmacol Sci 27:186-189]. So it appears that either, or both, CLOCK and its binding partner, BMAL1, play a central role in triggering the manic phase of bipolar disorder.

Furthermore, studies in humans reveal that variants, or polymorphisms, in Clock and its binding partner, Bmal1, are suspected to play a role in the frequency and severity of the manic phase of bipolar disorder. Thus, if one of these genes can be altered in mice, it will likely be possible to construct a mouse model in which to study the manic phase of bipolar disorder.

Using this information, a research team headed by Kole Roybal, mutated the Clock gene in mice and then subjected them to a series of behavioral experiments to determine the nature of this mutation in a living animal. These behavioral experiments on the mice expressing the mutated CLOCK protein showed that they exhibited locomoter hyperactivity, disrupted circadian rhythms, decreased sleep, reduced anxiety levels, less helplessness, and an increased preference for cocaine — similar to bipolar humans exhibiting mania. They also found that the experimental mice responded to treatment by lithium similarly to many humans with bipolar disorder.

Unfortunately, these CLOCK-mutated mice do not cycle between mania and depression, as bipolar humans do, so they are only useful for better understanding the underlying molecular aspects of mania/hypomania. Because they behave similarly to manic bipolar humans, and because they respond to lithium similarly to bipolar humans, these mice represent a good and useful beginning for understanding how to successfully manage bipolar disorder in humans.


Mania-like behavior induced by disruption of CLOCK by Kole Roybal, David Theobold, Ami Graham, Jennifer A. DiNieri, Scott J. Russo, Vaishnav Krishnan, Sumana Chakravarty, Joseph Peevey, Nathan Oehrlein, Shari Birnbaum, Martha H. Vitaterna, Paul Orsulak, Joseph S. Takahashi, Eric J. Nestler, William A. Carlezon, Jr., and Colleen A. McClung. Proceedings of the National Academy of Sciences (Mar 22, 2007) doi:10.1073/pnas.060962510407.

What can a clock mutation in mice tell us about bipolar disorder? by Joseph T. Coyle. Proceedings of the National Academy of Sciences (Apr 2, 2007) doi:10.1073/pnas.0701491104. (IMAGE)


    April 11, 2007

    I saw this doing the Stumble Upon and thought you’d like it. He sure has done a fine job.

  2. #2 JPS
    April 11, 2007

    This was interesting. As far as I know there is very little known about the mechanisms of how bipolar disorder works in the brain. Maybe this study will shed some light on the manic phases

    I remember reading how depakote is an effective treatment for bipolar disorder, but the mechanism of how it works is not really understood.

  3. #3 coturnix
    April 11, 2007

    Thank you for doing this. I only mentioned it a little bit when the press release first came out (but the link within may be of interest). I may just link to you instead of writing an entirely new post.

  4. #4 Bob O'H
    April 12, 2007

    Ah, thanks. I must admit, I skimmed through the Coyle commentary, and was scared off by the litany of gene and protein names. It’s a pity they couldn’t get a model with both phases: this work implies that there might be people who are permanently in the manic phase.

    I do wonder if things are going too far when proteins are being sponsored by drugs companies (Glaxo SmithKline sponsoring GSK3beta?).


  5. #5 David Harmon
    April 12, 2007

    But these mice represent a good and useful beginning for understanding how to successfully manage bipolar disorder in humans.

    (r)Amen! I had wondered whether mice could be afflicted with a BP model, but the importance of timekeeping genes here, suggests to me that getting the bimodality to work may be “just an engineering problem” after all. And of course, every step teaches us more, even before we start using the new Mutant Mouse Mania Model.

  6. #6 Deb
    April 17, 2007

    I came over from the carnival, and found this to be a fascinating article. I find seasons play a role in my patterns as well.

  7. #7 dan dright
    April 20, 2007

    I read this one (McClung) pretty carefully and it left me with a great many questions.

    I work with clock mutant flies, and have serious concerns with a number of things.

    1. I didn’t see any attention paid to overall metabolic changes in the clock mutants. Much of the results could be concievably explained by overall hyperactivity (in particular the forced swimming, learned helplessness, the elevated plus maze and the sucrose preference.) This is particularly vexing considering GSK beta is a major part of the glycolysis pathway, which is critical to ATP synthesis–where we get our energy from. Lithium is believed to act via GSK.

    2. The cocaine response was never normalized to appropriate levels for the mutants. Instead it is normalized to the WT animal. It is very possible that overall response curves are different to cocaine, depending upon mutant overall sensitivity phenotype.

    3. The cocaine response only becomes significant under tremendous levels of drug. At this point it is doubtful that we are looking at hedonia, but rather resistance to aversive effects. Cocaine at high doses is quite aversive.

    4. Clock is a dominant negative phenotype. In the absence of Clock altogether, a mutant phenotype is not seen, as a secondary protein (a PAS domain protein I believe) forms a heterodimer with BMAL, filling in for clock. Some controls, with clock null mice and clock + pas domain null mice.

    5. They don’t look at all at what is going on with the clock at all in these mice as a response to the assays. It would be nice to see what the lithium and assays are doing to the clock centers in WT and in mutants.

    That’s it for now. 🙂

New comments have been disabled.