DNA diversity in the brain: Somatic mutations reveal a neuron’s history

neurons somatic mutations
Neurons are more like snowflakes–no two alike–than anyone realized.

Walt Whitman’s famous line, “I am large, I contain multitudes,” has gained a new level of biological relevance in neuroscience.

As we grow, our brain cells develop different genomes from one another, according to new research from Harvard Medical School and Boston Children’s Hospital. The study, published last week in Science, provides the most definitive evidence yet that somatic (post-conception) mutations exist in significant numbers in the brains of healthy people—about 1,500 in each of the neurons they sampled.

The finding confirms previous suspicions and lays the foundation for exploring the role of these non-inherited mutations in human development and disease. Already, the researchers have found evidence that the mutations occur more often in the genes a neuron uses most. And they been able to trace brain-cell lineages based on mutation patterns.

“This work is a proof of principle that if we had unlimited resources, we could actually decode the whole pattern of development of the human brain,” says co-senior investigator Christopher Walsh, MD, PhD, the HMS Bullard Professor of Pediatrics and Neurology and chief of the Division of Genetics and Genomics at Boston Children’s. “These mutations are durable memory for where a cell came from and what it has been up to. I believe this method will also tell us a lot about healthy and unhealthy aging as well as what makes our brains different from those of other animals.”

Taking a history

Germline, or inherited, mutations have taken most of the blame for causing brain disorders such as Alzheimer’s, autism and schizophrenia. The role of non-inherited mutations in somatic cells has been murkier. Until recently, scientists didn’t even know if there were enough of them in the brain to matter.

“A lot of people have been asking whether somatic mutations contribute to neurodevelopmental and neurodegenerative diseases, but they couldn’t answer the question because of the limitations of technology,” says co-senior investigator Peter Park, PhD, associate professor of biomedical informatics at HMS.

Park, Walsh and team made headway toward answering these questions by combining DNA amplification, single-cell genome sequencing and rigorous data analysis techniques.

Stem cells in the developing brain (white dots at bottom) generate mature neurons (white teardrops at top). Image: Peter Wang and Matthew Johnson
Stem cells in the developing brain (white dots at bottom) generate mature neurons (white teardrops at top). Mutations can occur along the way, or later. (Image: Peter Wang and Matthew Johnson)

They studied a particular kind of somatic mutation called a single nucleotide variant or SNV—a single-letter change also known as a point mutation. Since SNVs may occur in just a few cells, or even just one, they can be hard to detect with whole-genome sequencing analyses that take the average of hundreds of thousands of cells. Sequencing individual cells brings the rare mutations to light.

The downside is that sequencing the genome of a single cell costs as much as sequencing an entire person’s genome. Working within this limitation, the researchers sequenced 36 neurons sampled from the cerebral cortices of three deceased people without brain disease, ages 15, 17 and 42. Each neuron, they found, contained about 1,500 SNVs.

For each person, the researchers compared the neurons’ genetic sequences to one another, and to the person’s heart cells. This allowed them to start figuring out which mutations were shared by which neurons and which could be found elsewhere in the body. They then conducted tests in several parts of the brain to see how many cells in each region carried the same mutations as the neurons they’d sequenced.

Grouping neurons based on their mutations provided clues about cell history, because neurons that shared mutations likely came from the same stem cell ancestor. If only a few cells shared a mutation, they probably belonged to a lineage that emerged more recently.

“If a mutation arose very early, it would be present both inside and outside the brain,” says Walsh. “Or if it arose later, it would be in certain parts of the brain and not others. We’re looking at a record of the series of cell divisions that generated the brain.”

Based on mutation patterns and location in the brain, some neurons’ lineages could even be traced to a specific day during embryonic development.

Use it and lose it

Somatic mutations have many causes. Ultraviolet light causes them in skin cells. Errors in DNA replication cause them in rapidly dividing cancer cells.

“What we found in the brain was neither of those things,” said Walsh. “We thought the dominant source of mutation would be faulty DNA replication and were surprised to find that instead, it’s faulty DNA expression.”

Park’s data analysis revealed that the genes with the most mutations tended to be the ones that were used most in the brain.

“People like to say about your brain, ‘use it or lose it,’” says Walsh. “Unfortunately, we found there’s a certain element of ‘use it and lose it.’ Whenever you turn genes on, they run the risk of being damaged because you have unwind the DNA and separate the two strands. The DNA becomes vulnerable while you’re using it. So every time we turn a gene on, it’s a risk.”

The findings make the team wonder whether somatic mutations accumulate with age and whether they might contribute to neurodegeneration. The researchers plan to investigate how, when and why new mutations arise in the brain, whether they are damaging or protective and whether mutation rates vary from person to person. They will also study other types of somatic mutations in the brain and other tissues.

As he helps to uncover the role of somatic mutations in the brain, Park has taken in stride the knowledge that our brains contain a multitude of genetic variation.

“I’m full of mutations but I’m walking around, pretty healthy,” said Park. “It just goes to show that there are a lot of things we don’t understand.”

Co-first authors of the study were Michael Lodato and Mollie Woodworth, postdoctoral researchers in pediatrics at Boston Children’s, and Semin Lee, postdoctoral researcher in biomedical informatics at HMS.

More on GenomeWeb, The Atlantic and the Howard Hughes Medical Institute website.