Our brain contains billions of cells called neurons, each connected to one another in a complex network, permitting us to think, move, and assimilate information about our environment. While many cells in our body can talk to one another, neurons are unique in that their communication relies on electrical activity. Much like how the flow of charged particles along electric cables facilitates information transfer, the flow of positive and negative ions across a neuron generates a pulse of electrical activity—neural activity—that is transmitted and replicated in subsequent neurons.

Mature neurons are also unique for their inability to divide. If we trip and scrape our skin, blood vessels and skin regenerate to fill the loss. In contrast, there are no newborn neurons present to replace ones lost in, say, Alzheimer’s disease (AD), which afflicts 5 million people in the U.S. every year [1]. In AD, neurons in a region of the brain regulating memory are highly vulnerable. The loss of these neurons produces dementia and memory loss in patients, but it’s unclear why these neurons are more vulnerable than others. To better understand the biology behind the disease, Dr. Lennart Mucke and his team at UCSF began assessing the extent of DNA damage in mouse models of Alzheimer’s and made a surprising discovery: neural activity itself can cause DNA damage.

Kamikaze neurons: neural activity generates double-stranded DNA breaks

Every cell in our body contains a copy of our DNA, the genetic material that serves as a blueprint for every living organism. When cells replicate, they need to copy their DNA.  This process often introduces errors, creating DNA damage that cells then need to repair. External forces, such as UV radiation, can also damage DNA.

DNA is made up of two strands of molecules intertwined in a double helix structure. DNA damage can cause either double-strand breaks or single strand nicks. Double-stranded breaks (DSBs) are the most severe type of DNA damage (Fig. 1). When a cell detects double-stranded breaks, it recruits a variety of proteins to the break site, some of which help to repair the break, and others with yet unknown functions. Monitoring the accumulation of such proteins (referred to hereafter as “DSB proteins”) is the easiest way for scientists to tell that a DSB has occurred.

Figure 1. Two strands of molecules comprise DNA in a double helix (left); a double-stranded break occurs when both strands are nicked, breaking the DNA apart into segments (right).

While assessing the extent of neuronal DNA damage in mice genetically predisposed to Alzheimer’s disease (AD mice), the researchers inadvertently discovered that placing normal mice in novel environments for a few hours led to an increase in the number of neurons with DSBs. These DSBs occurred selectively in regions of the brain regulating spatial exploration and new memory formation. After the authors returned the mice to their original cages, the DSBs were repaired over a 24-hour period. These results suggested that normal neural activity generated during learning about a new environment could stimulate DSBs.

Exploring a novel environment promotes a relatively gentle form of neural activity, and it is surprising that such minimal activity would produce DNA damage. To further investigate this, the scientists activated neurons in a more specific manner by covering the left eye of the mouse and exposing the right to visual cues. Because incoming sensory inputs are processed by brain regions on the opposite side of the body, the authors found a significant increase in the number of neurons containing DSB proteins only in the left visual cortex (corresponding to visual input from the right eye). These DSBs were repaired within 24 hours. They found no changes in DSB protein levels in the right visual cortex or in brain regions associated with touch. These results show that elevating neural activity in different neural networks increases DSB proteins, or DNA damage, specifically in neurons within that network.

β-Amyloid, exacerbating neuronal DNA damage

Intriguingly, compared to normal mice, when AD mice explored novel environments, they exhibited a greater increase in neurons with DSBs, and this change persisted even after their return to their original cages. In Alzheimer’s patients and mice, plaques, or abnormal deposits of β-amyloid (Aβ) protein fragments, appear throughout the brain and are thought to injure neurons. Aβ immediately arose as a prime suspect for stimulating and maintaining an increase in neuronal DSBs. To test this hypothesis, scientists isolated neurons from normal mice and added Aβ fragments to the neurons in culture. The addition of Aβ led to a three-fold increase in DSB protein levels, indicating that the presence of these deposits alone can cause DSBs.

However, recent research into the biological effects of Aβ in the brain has revealed that Aβ can raise the overall activity of neural circuits. For example, some of these AD mice have intermittent seizures, indicative of excessive neural activity. Since the authors previously showed that neural activity can generate DSBs, they needed to distinguish between the roles of Aβ and excessive neural activity. To do this, the authors manipulated activity in the presence of Aβ. They treated AD mice with the drug levetiracetam, which partially suppresses aberrant neural activity. After a month of treatment, researchers saw a reduction in neurons with DSBs in AD mice to levels of normal mice. In contrast, no changes in DSB protein levels were observed in normal mice treated with levetiracetam, suggesting that the additional activity in Alzheimer’s brains also contributes to DSBs.

But of the two—activity and Aβ—which makes the greater contribution to neuronal DSBs? This question led the researchers to add tetrodotoxin (TTX), a pufferfish toxin, into their cultured neurons. TTX prevents the flow of positive ions into neurons, thereby blocking neural activity in all cells. This complete loss of activity subsequently prevented Aβ from increasing DSB proteins. Thus, these data show that while Aβ can induce DSBs, it relies on the presence of neural activity (Fig. 2).

Figure 2. Normal neural activity from exploring the environment or visual stimuli can cause double-stranded breaks in DNA, but in healthy neurons, these are repaired within 24 hours (top pathway). In the presence of β-Amyloid, excessive baseline activity produces an excess of DSBs, which the neurons cannot heal within 24 hours (bottom pathway).

What’s next?

Considering that the words “DNA damage” have inherent negative connotations—indeed, scientists have always believed DNA damage to be detrimental to cells—this study revolutionizes the way scientists think about DSBs. By showing that DNA damage and repair occurs regularly with normal brain activity, Dr. Mucke and his team suggest that DSBs may not only be a normal part of brain function, but that they may also confer benefits to the neurons. Nevertheless, too much DNA damage may still prove harmful. In the face of excessive brain activity, as in AD, abnormally high levels of DSBs could interfere with normal development and contribute to neuronal injury and death.

Much more research is required to understand the role of DSBs in neurons and how they impact neuron function, but there are already many exciting avenues of research to pursue. One intriguing possibility is that these breaks in DNA could facilitate the development or maturation of our neurons as we learn and remember something new, giving us greater insight into how experience sculpts our own neural circuitry. Let’s hope that the scientists who continue studying this phenomenon take a leap of faith and assume that, in the long run, thinking won’t hurt them.

Andrea Yung is a graduate student in neuroscience at Harvard Medical School.

References

[1] Alzheimer’s Association (2013). What is Alzheimer’s? http://www.alz.org/alzheimers_disease_what_is_alzheimers.asp

[2] U.S. News Health (2008). Alzheimer’s Disease.

http://health.usnews.com/health-conditions/brain-health/alzheimers-disease

Further Reading

Brain DNA Damage Shown after Normal Learning Activity in Mice

http://www.huffingtonpost.com/2013/03/25/brain-dna-damage-learning-mice_n_2948180.html