Mutations accumulate in our brain cells as we age. Do they explain cognitive loss?

the aging brain - do DNA mutations in neurons account for cognitive loss?

Scientists have long wondered whether somatic, or non-inherited, mutations play a role in aging and brain degeneration. But until recently, there was no good technology to test this idea.

Enter whole-genome sequencing of individual neurons. This fairly new technique has shown that our brain cells have a great deal of DNA diversity, making neurons somewhat like snowflakes. In a study published online today in Science, the same single-neuron technique provides strong evidence that our brains acquire genetic mutations over time.

“It’s been an age-old question as to whether DNA mutations can accumulate in neurons and whether they are responsible for the loss of function that the brain undergoes as we get older,” says Christopher A. Walsh, MD, PhD, chief of the Division of Genetics and Genomics at Boston Children’s Hospital and co-senior investigator on the study. “It hasn’t been possible to answer this question before, because we couldn’t sequence the genome of a single cell, and each mutation accumulated is unique to each cell.”

Interestingly, the researchers found that mutations build up even in brain cells that don’t divide, missing out on a prime opportunity for mutations to occur. They also showed that people with disorders causing early brain degeneration acquire mutations in their neurons at a faster rate.

Testing neurons one by one

The researchers obtained postmortem brain samples from the NIH NeuroBioBank. They tested DNA from 161 individual neurons, taken from 15 neurologically normal people of different ages (ranging from 4 months to 82 years) and nine people with Cockayne syndrome or xeroderma pigmentosum, The latter are genetic disorders that cause premature aging and early brain degeneration.

The genosenium - patterns of mutation in normal aging brains and two genetic diseases of early aging and neurodegeneration.
Three mutation signatures (click to enlarge): This graph of all 161 neurons tested, from people of different ages, shows the quantity and type of mutations they carried. From left to right are neurons from normal controls, people with Cockayne syndrome (CS) and people with xeroderma pigmentosum (XP), by ascending age in each group. Color indicates the type of mutation: Signature A (purple) correlates closely with aging; Signature B (orange) is apparent soon after birth;  Signature C (blue) is related to oxidative DNA damage. Mutations are also grouped by brain location: PFC, prefrontal cortex; DG, dentate gyrus of the hippocampus. Neurons from the older controls and the two disease groups had the highest numbers of mutations. (Reprinted with permission from MA Lodato et al., Science December 7, 2017, DOI: 10.1101/221960)

The team’s cutting-edge methods allowed for detection of DNA mutations as tiny as single-letter changes in each neuron. This required amplifying each cell’s genome — by generating a multitude of copies — and analyzing a boatload of data.

“Because many experimental artifacts arise during the single-cell experiments, a new computational method that can distinguish true mutations from the experimental noise was critical to the success of the project,” says Peter J. Park, PhD, of Harvard Medical School’s Department of Biomedical Informatics (DBMI), the paper’s other co-senior author.

The neurons tested came from two brain areas implicated in age-related cognitive decline: the prefrontal cortex, the part of the brain most highly developed in humans, and the dentate gyrus of the hippocampus, a focal point in age-related degenerative conditions like Alzheimer’s.

In neurons from neurologically normal people, the number of genetic mutations increased with age in both brain areas. However, mutations accumulated at a higher rate in the dentate gyrus. The researchers think this may be because these neurons have the ability to divide, unlike their counterparts in the prefrontal cortex.

Neurons from people with Cockayne syndrome and xeroderma pigmentosum showed an increase in mutations in the prefrontal cortex over time — more than double the normal rate. With help from collaborators at WuXi NextCODE, the researchers also showed that the portions of the genome that neurons used the most gained mutations at the highest rate.

The aging genome: How’s your genosenium?

To capture the idea of a gradual, inevitable accumulation of mutations contributing to brain aging, the researchers coined the term “genosenium” — a mashup of “genome” and “senescence” or “senility.”

Metaphor for the aging brain, losing cognitive function as it acquires NDA mutations

The mutations fell in three categories. “We were able to take all the mutations we found and use mathematical techniques to deconstruct them into different types of DNA changes,” says Michael Lodato, PhD, one of the six co-first authors on the paper. “It’s like hearing an orchestra and teasing out the different instruments.”

One category of “clocklike” mutations was strictly aging-related. These mutations accumulated like clockwork in both brain areas and were independent of disease status. Another type of mutation did not correlate with age, except in the dentate gyrus, where mutation numbers did increase over time.

A parallel with cancer?

The third type was associated with oxidative damage to DNA and faulty DNA repair. These mutations increased with age and were seen in high numbers in Cockayne syndrome and xeroderma pigmentosum neurons, and to a lesser extent in normal neurons.

“This last finding convinced me I need more anti-oxidants,” quips Walsh. “Overall, it raises a question as to whether neurodegenerative diseases are like cancer, relating ultimately to DNA mutation.”

It also raises a question about whether age-related mutations contribute to everyday “senior moments,” something a study of postmortem brain cells couldn’t capture.

The researchers are now turning their sights on neuronal mutation patterns in other neurodegenerative disorders. “The technology we used can be applied to any degenerative disease of the brain,” says Walsh.



Michael Lodato, Rachel Rodin and Michael Coulter of Boston Children’s and Craig Bohrson, Alison Barton and Minseok Kwon of Harvard Medical School’s DBMI were all co-first authors on the study. Other coauthors were: Maxwell Sherman, Carl Vitzthum and Lovelace Luquette of HMS DBMI; Chandri Yandava, Pengwei Yang and Thomas Chittenden of the WuXi NextCODE Advanced Artificial Intelligence Research Laboratory; and Nicole Hatem, Steven Ryu and Mollie Woodworth of Boston Children’s.

The study was supported by the National Institutes of Health (K99 AG054749 01, F30 MH102909, 1S10RR028832-01, T32HG002295, U01MH106883, P50MH106933, R01 NS032457, U01 MH106883), the Harvard/MIT MD-PHD program, the Stuart H.Q. and Victoria Quan Fellowship in Neurobiology, the Allen Discovery Center program through The Paul G. Allen Frontiers Group and the Howard Hughes Medical Institute.