The past few decades have seen intensive efforts to find the genetic roots of neurological disorders, from schizophrenia to autism. But the genes singled out so far have provided only sketchy clues. Even the most important genetic risk factors identified for autism, for example, may account for only a few percent of all cases.

Much frustration stems from the realization that the key mutations elevating disease risk tend to be rare because they are less likely to be passed on to offspring. More common mutations confer only small risks (although those risks become more significant when calculated across an entire population). There are several other places to look for the missing burden of risk, and one surprising possible source has recently emerged—an idea that overturns a fundamental tenet of biology and has many researchers excited about a completely new avenue of inquiry.

Accepted dogma holds that—although every cell in the body contains its own DNA—the genetic instructions in each cell nucleus are identical. But new research has now proved this assumption wrong. There are actually several sources of spontaneous mutation in somatic (nonsex) cells, resulting in every individual containing a multitude of genomes—a situation researchers term somatic mosaicism. “The idea is something that 10 years ago would have been science fiction,” says biochemist James Eberwine of the University of Pennsylvania. “We were taught that every cell has the same DNA, but that’s not true.” There are reasons to think somatic mosaicism may be particularly important in the brain, not least because neural genes are very active.

A paper published on April 28 in Science by a group founded two years ago—the Brain Somatic Mosaicism Network (BSMN)—outlines a research agenda for using new technologies to explore the genetic diversity found in each cell and to investigate what links, if any, tie such mutations to a variety of neurological conditions. “The field was abuzz with interest in exploring mosaicism, but there was no money,” says Thomas Lehner, director of the Office of Genomics Research Coordination at the National Institute of Mental Health, which is now devoting $30 million in funding to the BSMN over the first three years, two of which have elapsed.

The consortium consists of 18 research teams at 15 U.S. institutions with access to repositories of postmortem brain tissue taken from healthy people and others with schizophrenia, autism, bipolar disorder, Tourette’s syndrome or epilepsy. Each team is tackling different samples. “There’s a lot of new technology application and development involved, and a ton of data that will become a resource,” Lehner says. “We also wanted to understand if there’s an association with new technology, so we encouraged researchers to include brain banks of individuals with various neurological conditions.” 

Studies that preceded the consortium have confirmed mosaicism is commonplace. One report estimated there may be hundreds of changes in single letters of genetic code (single-nucleotide variants, or SNVs) in each neuron in the mouse brain. Another found more than 1,000 in human neurons. These findings suggest somatic mosaicism is the rule, not the exception, with every neuron potentially having a different genome than those to which it is connected. A primary cause of somatic mutations has to do with errors during the DNA replication that occurs when cells divide—neural progenitor cells undergo tens of billions of cell divisions during brain development, proliferating rapidly to produce the estimated 80 billion neurons in a mature brain. The image of each cell carrying a carbon copy of the genetic material of all other cells is starting to fade—and for good reason. Genetic sequencing does not normally capture the somatic mutations in each cell. “You get a sort of average of the person’s genome, but that doesn’t take into account any brain-specific mutations that might be in that person,” says lead study author Michael McConnell of the University of Virginia.

A 2012 study found somatic mutations in the brains of children with hemimegalencephaly, a developmental disorder in which one hemisphere is enlarged, causing epilepsy and intellectual disability. The mutations were found in brain tissue but not always in blood or in cells from unaffected brain areas and in a small fraction (around 8 to 35 percent) of cells from affected areas. Such studies, showing somatic mutations can cause specific populations of cells to proliferate and lead to cortical malformations, have researchers wondering whether somatic mutations may also play a role in more complex conditions.

Mature neurons stop dividing and are among the longest-living cells in the body, so mutations will stick around in the brain. “In the skin or gut, cells turn over in a month or week, so somatic mutations aren’t likely to hang around unless they form cancer,” McConnell says. “These mutations are going to be in your brain forever.” This could alter neural circuits, thereby contributing to the risk of developing neuropsychiatric disorders. “In psychiatric disease, we don’t know that much yet, and that’s largely the goal [to find an answer],” McConnell says. “It’s a good hypothesis, but it’s going to require this big, multiteam effort to really address it.” To investigate, the consortium will sequence brain DNA from control and patient samples. “Before you can get to your destination, you have to have a map, and this is going to help build that map of somatic mutations that have potential for influencing neural functioning and disease,” says Eberwine, who was not involved in the new research. “So this consortium is critically important for neuroscience.”

One question to be explored is whether genes associated with a brain disorder may harbor somatic mutations. The fact that specific genes explain just a small proportion of cases may be because researchers have been looking only in the germ line (sex cells), McConnell says. “Maybe the person doesn’t have the mutation in their germ line, but some percentage of their neurons have it.” Somatic mosaicism may also contribute to neural diversity in general. “It might explain why everybody’s different—it’s not all about the environment or genome. There’s something else,” says neuroscientist Alysson Muotri of the University of California, San Diego, who is not part of the consortium. “As we understand more about somatic mosaicism, I think the contribution to individuality as well as the spectrum [of symptoms] you find in, for example, autism, will become clear.”

Somatic mutations can occur in multiple circumstances. They may emerge during DNA replication or from DNA damage (caused by free radicals or environmental stresses), combined with imperfect repair machinery. In addition to SNVs, mutations known as indels, involving insertions and deletions of small DNA sequences (typically tens of nucleotides), also occur frequently. Larger, rarer mutations include structural changes in chromosomes, in either the form of gains or losses of whole chromosomes or copy number variants (CNVs), in which the number of repetitions of long chunks of DNA (covering multiple genes) is altered. Within genomes, there are also mobile genetic elements that act almost like parasites, jumping around or making copies of themselves and inserting themselves elsewhere in the genome, seemingly to ensure their survival. These strange entities are an active field of research in their own right: they are important here because they can cause somatic mutations, including a type known as mobile genetic element insertions, or MEIs. They are switched on in the same way as genes involved in producing new neurons, making them especially active in the brain during development.

The paper outlines three methods for studying these mutations. The first involves using technologies to sequence a whole genome from bulk brain tissue. This technique can detect many variants, but the rarest types are diluted by the mass of cells in bulk tissue. “Large CNVs and mobile elements are much more difficult to detect in bulk tissue than SNVs,” McConnell says. Also, this method cannot reveal how mutations vary between cell types. This can be partly solved using a technique known as sorted pools, which sorts out neurons from other unwanted cell types. The most important recent advance that will aid the consortium, however, is the advent of technologies that allow the genomes of individual cells to be sequenced. “By going into single cells, we can compare [what we find] to the neighboring cell and say, ‘Aha, they’re different!’ That’s the advance that allows us to really move forward,” Muotri says. “I’m very excited—this is the beginning of something completely new in biology and neuroscience.”

The project is funded until 2020 and will make all data publicly available—and for some results, that should be in 12 to 24 months. “Around 10,000 sequencing data sets will be generated, and we’ll be making that available in a database for the scientific community to dig in more deeply,” McConnell says. There are also plans to collaborate with other NIMH initiatives, including BrainSpan, which maps gene expression during brain development, and psychENCODE, which is mapping the brain epigenome (environmentally driven modifications of DNA that influence gene activity without changing the genetic code). “This is supposed to initiate an important area of research,” Lehner says. “We hope it will give us a landscape of mosaicism in the brain and insights into the contribution of mosaicism to mental disorders, but I don’t expect to have all the answers.” Such insights may ultimately lead to the discovery of new genetic targets for treating a range of hard-to-treat disorders.

“This is exploratory research; we’re learning about the phenomenon,” Muotri says. How important it will be is not clear at this stage, but “by figuring out how it works, we may reveal new therapeutic opportunities.”