Too Much Thinking Can Cause DNA Damage In The Brain?
    By Jennifer Wong | April 25th 2013 04:49 PM | 8 comments | Print | E-mail | Track Comments
    About Jennifer

    My column covers the latest primary research discoveries in the life-science discipline. Much of what is reported here are considered discoveries...

    View Jennifer's Profile

    Neuroscientists have long demonstrated that neuronal connections in the brain can be strengthened with neuronal activity in the process known as neuroplasticity, and that brain training can be the ideal remedy to sharpen the human mind and to slow down the progress of neurodegeneration. However, recent studies revealed that too much thinking can actually be detrimental to the brain, causing profound DNA damage often dubbed as the DNA double-stranded breakages (DSBs).

    DSBs are identified by the accumulation of gH2A.X histone- a recruiter of the DNA-repair machinery- at the site of breakage, and are previously thought to be caused only by cell stress.

     According to the study by Suberbielle et al. published in this issue of Nature Neuroscience, DSBs are also produced when the brain is at work (during exploration tasks), or when neurons are aberrantly activated during an epileptic seizure or by artificial stimulation by optogenetic techniques. The authors further showed that the DSBs are not produced as a response to stress hormones, but are rather produced by the normal neuronal activation of the NMDA glutamate receptor; a key receptor involved in neuroplasticity.

    The effect of neuronal activity on DSB production is further aggravated by the presence of amyloid-beta plaques in the brain of Alzheimer’s disease patients; an effect induced by aberrant neuronal activity or seizures associated with the disease. While DSBs are dangerous DNA lesions that can result in neuronal cell death and the loss of brain function, the DSBs are shown to be repaired within 24 hours, begging the question as to whether DSB represents a crucial and yet unknown genetic event in memory and cognition.

    Indeed, this discovery has far-reaching implications in our understanding of the neurological consequences of brain training. With brain-training games supporting a growing multi-million dollar industry, the study by Suberbielle et al. is forcing scientists and brain-training consumers to question whether brain-training could potentially have negative consequences in the brain. The study could also explain why there is very little consensual scientific evidence showing the effectiveness brain-training games on improving the human mind.  

    Lastly, as brain training is shown as an effective a remedy against the onset of Alzheimer’s disease, the aggravation of activity-induced DSBs in the Alzheimer’s brain is a ground-breaking discovery fostering questions as to whether brain-training is a safe remedy for Alzheimer’s disease. Further research into the role of activity-induced DSBs in memory and cognition would be crucial to answer this pressing question.


    Suberbielle et al. Physiologic brain activity causes DNA double-strand breaks in neurons, with exacerbation by amyloid-β. Nature Neuroscience. 2013 Mar 24. doi:10.1038/nn.3356


    Gerhard Adam
    ... and what exactly is the scientific definition of "too much thinking"?
    Mundus vult decipi
    Thanks for your feedback. "Too much thinking" is simply a lay expression indicating extensive use of the brain's cognitive capacity to do mental tasks. The title here is quite creative and is geared towards a lay audience- so I don't want to complicate this.  I get your point, but let's not dwell in the specifics of scientific terminology in this discussion. I would also really appreciate it if you could refrain from further discussion in the "wording" matter- and leave the space for other scientific discussion. I hope you understand. Many thanks. 
    Are there cases in which people have been hospitalized because of this?
    Probably not....but I can't comment since this study is done only in rodents.
    But remember that the activity-induced DNA breakages are normally repaired within 24 hours, at least in the normal brain. The DNA repair is less effective in the Alzheimer's brain. 
    This also brings up an interesting point that perhaps the normal brain needs enough rest between intervals of cognitive allow activity-induced DNA breakages to be repaired. Perhaps too much cognitive work and too little rest might be detrimental to the brain. Something to think about...

    The following is not a "wording" question, it is a question of experimental conditions: are you referring to activities in which a significant amount of learning is repeatedly called for, over a significant period of time? 
    In principle, there is no reason to expect that learning would necessarily be cost-free in terms of its effect on other brain activities and processes. Such trade-offs are frequently encountered in, for example, semiconductor design. It should not be surprising to discover that learning has side effects.
    I wouldn't say DNA damage is the result of significant amount of learning...the author is suggesting that any level of learning activity (no matter how hard or easy) can cause DNA damage. But because the DNA get's repaired within 24 hours when the brain is not learning, it's tempting to speculate that learning without rest, say for an extended period time, could cause the accumulation of DNA damage in the brain.
    I would really like to correct something here. "Physiologic brain activity" is neither synonymous with nor descriptive of "too much thinking", and so is scientifically misleading.

    Regardless of what wording is chosen, I would view the results of this study with caution. For one, the vast majority of the assays conducted by the authors involve measurements of phosphorylation on histone 2A variant X (γH2A.X). Since γH2A.X is regularly used as a marker of DNA double-stranded breaks induced by radiation or occurring in cancer cells, the authors infer that novel exploration and other tasks that increase neuronal activity are causing DNA damage. But this of course assumes that the formation of γH2A.X is a reliable marker of only DNA damage in neurons, whereas this site may be phosphorylated in the absence of DNA damage (e.g.

    Double-stranded breaks are only definitive identified in a single case. The authors had groups of mice either stay in their home cages or enter a novel environment (n = 6-9 per group). Immediately afterwards, dentate gyrus tissue was extracted and cell nuclei were run by single-cell electrophoresis on the "comet assay". Using this assay, nuclear DNA generally remains clustered together in a little ball, but following double-stranded breaks, a comet-like smear forms behind the clustered DNA, representing DNA fragments which have exited the nucleus. The percentage nuclei forming a comet tail increased from ~15% to ~35% (roughly the same effect was seen following low-dose gamma irradiation).

    The most apparent issue with this latter finding is that it may not be specific to DNA breaks occurring in vivo . Note that roughly 15% (!) of cells from the dentate gyrus had comet tails in the control condition, potentially suggesting a high level of background false positives. The News and Views piece on this article ( points out that neural activity (and/or γH2A.X binding) might alter the conformation of nuclear DNA, making it more susceptible to mechanical damage caused by the assay . They also note that this interpretation is testable: if DNA damage really did increase, then other, definitive components of the DNA repair machinery should also have been recruited.

    The alternate is to assume that on a regular (e.g. hourly) basis, normal brain activity is constantly churning out double-stranded breaks at a very high rate. These breaks can only be repaired in non-dividing cells by a highly error prone process. In other words: DNA damage should be accumulating at a rapid rate in the brain. Therefore, the prior plausibility of this result appears to be pretty low.

    The "brain activity" that the paper here is referring mainly to the "neuronal activity" in the hippocampus- which is the essentially the center for learning and memory formation. In lay terms, this "brain activity"  can really be interpreted as "thinking". The lay term of  "too much thinking" simply refers to the prolonged mental exercise (physiological brain activity in the hippocampus), and little rest to allow the "DNA damage" to be repaired (which the paper suggests to be within 24 hours).  
    About the technique of this raise a good point. But if the gH2A.X binding is an artifact, then how do you explain why there is an increase in gH2A.X binding in the AD brain, or changes in response to brain activity? You wouldn't expect to see any clear difference if the assay only reflects an artifact? Seems a bit too coincidental. But you're right- maybe they should at least use another technique to confirm the presence of DNA breaks.