This story originally appeared on the Alzheimer Research Forum.
November 6, 2013. The word "mosaic" conjures up images of colorful tiles artfully arranged. For neurobiologists, the word may invoke the genetic hodgepodge that is the human brain. In the November 1 Science, researchers led by Fred Gage at the Salk Institute for Biological Studies, La Jolla, California, report that the genetic makeup of our most vaunted organ is actually a mosaic of genetically distinct neurons. They found scads of genomic changes in single cells from individual brains. Neurons with different genomes are bound to have different phenotypes, though what this means for the health of the brain is unclear.
The scientists had previously found that neurons can lose or duplicate entire chromosomes and that small retrotransposons frequently copy and paste themselves within the genome of a given brain cell (see Rehen et al., 2005; Muotri et al., 2005). First author Michael McConnell, now at the University of Virginia in Charlottesville, and colleagues characterize copy number variations (CNV) as a third type of genomic variability in neurons. These kilobase- to megabase-sized chunks of inserted, duplicated, or deleted DNA are common in the human body, including in the central nervous system (see O’Huallachain et al., 2012). Until now, most CNV assays were conducted with batches of cells, leaving scientists to wonder how often these alterations occurred at the single-cell level.
McConnell and colleagues address this question by sequencing the whole genomes of induced pluripotent stem cell (iPSC)-derived neurons from healthy people and frontal cortex neurons taken postmortem from young adults (see Vanneste et al., 2009; Navin et al., 2011). The researchers amplified the genome, then identified regions that seemed to contain more or less than the normal amount of DNA.
The results paint a complex picture. Out of 40 iPSC-derived neurons from three different cell lines, 13 contained one or more genome differences compared to the donor’s bulk DNA. Some contained gross changes, such as seven chromosome additions and four losses. Most contained at least one CNV and one cell had five. Together, these modifications far exceeded those found among fibroblasts or neural progenitor cells.
The genetic abnormalities in the postmortem human brain tissue were similar. Among 110 frontal cortex neurons taken from three normal people aged 20–26, 41 percent had at least one CNV. They ranged in size between 2.9 and 75 megabases, enough to cover several genes. While many cells contained few if any CNVs, 15 percent accounted for 73 percent of the variation. The CNVs in these cells spanned the entire genome, but seemed to be enriched in telomeres, the caps that protect chromosomes during DNA replication. Deletions were twice as common as duplications, though one person’s neurons showed more of the latter.
“Because these are such large changes, with a lot of genes being affected, they are unlikely to all be benign,” said Michael Snyder, Stanford University, California, who was not involved in the project. In fact, the authors were conservative when estimating the number of CNVs, meaning some likely fell below the detection threshold, Snyder speculated. “Presumably this is the tip of the iceberg for what is actually happening. The genome isn’t as constant as we expected,” he said.
Researchers are unsure what the findings mean. Gage and colleagues hypothesize that the genetic mosaicism allows for a broader range of phenotypes than would occur if all neurons had the same genome, and could help explain why animals and people with the same genetic makeup, for example identical twins, behave differently from one another (see Singer et al., 2010). Others suggest that mosaicism predisposes to age-related disorders of the central nervous system (see Fischer et al., 2012). Copy number variations carried through the germline cause Down syndrome, Alzheimer’s and Parkinson’s diseases, but would extra gene copies in single somatic cells be enough to trigger neurodegeneration? It is possible, agreed Snyder and McConnell, but more research will be needed to find out.
While whole chromosomal additions or deletions happen early in development during mitosis, smaller structural changes such as CNVs can happen over a neuron’s lifetime, McConnell told Alzforum. One cause may be double-strand DNA breaks resulting from oxidative stress, thought to be a risk factor for neurodegenerative diseases; repair mechanisms can introduce CNVs into the genome when they incorrectly fix double-strand breaks, McConnell explained. Interestingly, researchers from Lennart Mucke's lab at the Gladstone Institutes in San Francisco, California, recently reported that electrophysiological activity causes double-strand breaks in DNA. This could explain why neurons are likelier to have CNVs than their progenitors, the authors wrote.
McConnell pointed out that abnormalities in neurons are likely to have more consequences than those in other cells. Unlike most somatic cells, neurons do not divide, so changes accumulate and stick around. Neurons are also highly interconnected, so a change in one cell could modify a network. The authors plan to determine if CNVs alter neural circuits.—Gwyneth Dickey Zakaib.
McConnell MJ, Lindberg MR, Brennand KJ, Piper JC, Voet T, Cowing-Zitron C, Shumilina S, Lasken RS, Vermeesch JR, Hall IM, Gage FH. Mosaic copy number variation in human neurons. Science. 2013 Nov 1;342:632-637. Abstract