Saturday, May 9, 2009

Differential reactivity to emotional inputs and BDNF

Someone on the 23andme forum posted the following paper abstract:­hp?title=Rs6265
1: Am J Med Genet B Neuropsychiatr Genet. 2009 Apr 22. [Epub ahead of print]
BDNF, relative preference, and reward circuitry responses to emotional communication.

Gasic GP, Smoller JW, Perlis RH, Sun M, Lee S, Kim BW, Lee MJ, Holt DJ, Blood AJ, Makris N, Kennedy DK, Hoge RD, Calhoun J, Fava M, Gusella JF, Breiter HC.

Motivation and Emotion Neuroscience Collaboration (MENC), Athinoula A. Martinos Center in Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts.

Brain derived neurotrophic factor (BDNF) regulates neural development and synaptic transmission. We have tested the hypothesis that functional variation in the BDNF gene (Val66Met polymorphism, rs6265) affects brain reward circuitry encoding human judgment and decision-making regarding relative preference. We quantified relative preference among faces with emotional expressions (angry, fearful, sad, neutral, and happy) by a keypress procedure performed offline to measure effort traded for viewing time. Keypress-based relative preferences across the ensemble of faces were mirrored significantly by fMRI signal in the orbitofrontal cortex, amygdala, and hippocampus when passively viewing these faces. For these three brain regions, there was also a statistically significant group difference by BDNF genotype in the fMRI responses to the emotional expressions. In comparison with Val/Met heterozygotes, Val/Val individuals preferentially sought exposure to positive emotions (e.g
., happy faces) and had stronger regional fMRI activation to aversive stimuli (e.g., angry, fearful, and sad faces). BDNF genotype accounted for approximately 30% of the variance in fMRI signal that mirrors keypress responses to these stimuli. This study demonstrates that functional allelic variation in BDNF modulates human brain circuits processing reward/aversion information and relative preference transactions. (c) 2009 Wiley-Liss, Inc.

This is interesting to me. I'll get to why further along in the post, but first I've got some background to fill in:

I've been looking into how the limbic system, which includes the amygdala and hippocampus, is involved in processing and generating emotional communication (facial expression, body language, tone, that sort of thing) and physiological responses to such communication.

It seems to me like for some people, including me, the output of the emotional communication detectors seem to be wired to physiological responses with a much higher gain than they are for other people, like my brother. Put him in a room with an angry person and he'll be aware of it in an abstract sort of way, but physiologically seem unaffected. In the same situation my heart rate will be going crazy and I'll feel a strong urge to run for the exit, even if I know intellectually that the angry person isn't a threat and that I have nothing to do with whatever it is he's mad about. I've put a fair bit of effort into trying to understand this discrepancy.

The limbic system is designed to keep us from getting killed and/or eaten. It responds quickly, but it's not big on context or discernment. It's perfectly capable of causing sympathetic activation, releasing epinephrine (aka adrenaline) and all those other fun stress response chemicals, before the cortex has an opportunity to weigh in on the matter. Anyone who has rounded a corner while hiking and found themselves jumping backwards with their heart racing from what the cortex easily identifies a fraction of a second later as a stick knows this.

Your cortex can then either work to get the limbic system to settle down or make the situation worse depending on how you end up thinking about the event. Even best case though, it takes time for all those nasty stress chemicals to dissipate. So, in the example of me and my brother in the room with an angry person, I don't believe that the difference is due to a difference in how we're thinking about things.

These sorts of differences in reactivity can also be related to a difference in life experience. The limbic system learns and adapts over time based on earlier outcomes -- what circumstances turned out to be dangerous or not. Once you burn your hand on a stove you find yourself acting wary thereafter around stoves even without conscious thought. This is why burning your hand is so much more effective than someone repeatedly telling you to be careful around the stove (limbic vs. cortical processing). However, based on stories my mother told me, the difference between my and my brother's reactions to aversive emotional output in our general vicinity was apparently present from infancy, so I don't think limbic training is enough to explain it.

All this leaves me looking for a genetic basis for this difference. I may have found one potential piece of the explanation in rs1801252, a SNP in the ADRB1 receptor. If I'm interpreting the reference on the SNPedia page right, the fact that I'm homozygous for the atypical version of this (GG) would lead to a more drastic than normal cardiac response to a given level of epinephrine. However, it wouldn't affect the aspects of sympathetic nervous system activation other than those mediated by the beta 1 adrenergic receptors, and I don't get the impression that my brother is experiencing those other effects either. I think he's not ending up with as strong a signal from the emotional state detector system and/or experiencing a lesser limbic reaction to it in the first place.

This brings us back to why this study seems like an interesting potential lead. They're showing a pretty strong difference in limbic response to facial expression based on differences in genotype for BDNF. I've got the version of that SNP that seemed to have the stronger response (CC = Val/Val). I wonder what version of this my brother has? I guess I've gotta get him genotyped to find out...

There's obviously more to it than that. About 68% of the population are also CC on this SNP. If 68% of people have as strong a physiological reaction to emotional output in their vicinity as I do most of them hide it well. Elaine Aron has written a series of books about "highly sensitive people"(HSPs), which I think corresponds fairly well to what I've been talking about. She calls this trait Sensory-Processing Sensitivity (SPS) and estimates 15 to 20% of people to have it. I wonder how well her concept of the SPS trait would correlate with the BDNF genotype and fMRI testing results described in this paper.

I also wonder about how all this relates to autism spectrum disorders. It seems like part of what's going on there is that either the parts of the limbic system detecting the emotional state of others doesn't work well, or the degree to which that information impacts and/or is accessible to the affected individual is muted enough so as to cause them problems. This seems potentially a more extreme degree of the same effect that I've been talking about. Perhaps it's all one spectrum: people like me and other HSPs at one end, people like my brother who match social expectations on this issue in the middle, and people with various degrees of autism spectrum issues at the other end. I could be totally off base here -- I haven't looked into the literature on this issue much at all -- but I think it's an interesting area for study.


  1. Interesting post!
    I'm attending the conference of the Canadian Association for Neuroscience this week and today I saw a really interesting poster that raises interesting concerns with the interpretation of fMRI data, though. Essentially, it said that energy use by brain cells (which is related to oxygen use and blood flow) is not necessarily correlated with activity.

    Either way, interesting story about BDNF.

  2. Thanks for the comment Scientific Chick. Did they by any chance have a suggestion for what changes in energy use by brain cells in a particular part of the brain DOES mean if it isn't necessarily correlated with activity?

    I just potentially got a lead on getting access to a research fMRI, so I want to have *some* hope that it could tell me something useful...