Some people are human genetic chimeras which have cells from two different fraternal twins which fused into a single human during early embryonic development. Human chimeras are thought to be rare. But chimeras have produced some amazing medical stories such as the woman who failed a genetic test to prove she was the mother of her children. Turns out her ovaries came from a fraternal twin. Now some research on large genetic variations in human brains show that even people who started with a single genome at the embryo stage end up with a lot of genetic diversity between neuron cells taken from the same brain and lower but still substantial amounts of genetic diversity in skin cells.
It was once thought that each cell in a person's body possesses the same DNA code and that the particular way the genome is read imparts cell function and defines the individual. For many cell types in our bodies, however, that is an oversimplification. Studies of neuronal genomes published in the past decade have turned up extra or missing chromosomes, or pieces of DNA that can copy and paste themselves throughout the genomes.
The only way to know for sure that neurons from the same person harbor unique DNA is by profiling the genomes of single cells instead of bulk cell populations, the latter of which produce an average. Now, using single-cell sequencing, Salk Institute researchers and their collaborators have shown that the genomic structures of individual neurons differ from each other even more than expected. The findings were published November 1 in Science.
"Contrary to what we once thought, the genetic makeup of neurons in the brain aren't identical, but are made up of a patchwork of DNA," says corresponding author Fred Gage, Salk's Vi and John Adler Chair for Research on Age-Related Neurodegenerative Disease.
One implication: Research that uses identical twins to control for genetic influences on cognitive function understate the impact of genetic differences because lots of genetic differences exist in just a single brain. It also raises interesting possibilities: Maybe some geniuses (and some really low intelligence people) are the result of some copy number variations that were created when their brains were at early stages of development.
If copy number variations (CNVs) are creating differences in cognitive function in a single brain then that makes genetic research on the brain much harder. Every neuron tells its own genetic story. Also, getting one's whole genome sequenced becomes much harder and more expensive. Cells isolated from different locations in the body will tell a different tale. Sequence enough cells and it should be possible to identify which genetic variations arose after the initial fertilized egg. That will be useful. But all sorts of genetic variations in neurons or in other cells could make big impacts on local tissue function.
This result demonstrates why we need the ability to sequence a whole genome for just $100. It is easy to conceive of reasons why you could want to sequence the full genome for cells from 100 parts of your body. Local genetic variations could give you an easily upset stomach, a sleep disorder, or a skin condition. Lots of neurons have genetic copy number variations (CNVs) that didn't come from parents.
In the study, led by Mike McConnell, a former junior fellow in the Crick-Jacobs Center for Theoretical and Computational Biology at the Salk, researchers isolated about 100 neurons from three people posthumously. The scientists took a high-level view of the entire genome---- looking for large deletions and duplications of DNA called copy number variations or CNVs---- and found that as many as 41 percent of neurons had at least one unique, massive CNV that arose spontaneously, meaning it wasn't passed down from a parent. The CNVs are spread throughout the genome, the team found.
The miniscule amount of DNA in a single cell has to be chemically amplified many times before it can be sequenced. This process is technically challenging, so the team spent a year ruling out potential sources of error in the process.
Skin cells contain substantial genetic differences as well.
Interestingly, the skin cells themselves are genetically different, though not nearly as much as the neurons. This finding, along with the fact that the neurons had unique CNVs, suggests that the genetic changes occur later in development and are not inherited from parents or passed to offspring.
This study underscores the need to be very careful when creating stem cell lines for therapeutic purposes. Multiple cells will need to be isolated from different locations in the body, separately grown up into many cell lines, and then each cell line tested to determine whether it has too many harmful mutations to be suitable for growing replacement organs or for doing cell therapy to repair some organs.
Some of the CNV differences will be detectable using genetic tests that are much cheaper than genome sequencing. I also expect we will see the development of tests of messenger RNA (what DNA gets transcribed into before getting translated into proteins) to look for signs that some genes are expressed at much higher or lower levels due to CNVs.
Update: Chimeras could be avoided by use of in vitro fertilization. Only implant one embryo per cycle.
But how to control CNVs? That seems much harder. We need to know why they form more in neurons. Perhaps the formation of at least some of them is under a form of genetic regulation. One way to get a clue: sequence the genomes of twins. Also, sequence the genomes of multiple generations of a family. Do CNVs expand and contract across generations?
A recent story in Nature, Genome hacker uncovers largest-ever family tree, describes a family tree of over 13 million people. What is exciting about this. If substantial number of the living people in this family tree were sequenced we'd be able to see how fast single letter mutations and larger mutations happen across generations. How often do inherited CNV changes happen? We could know.
|Share |||Randall Parker, 2013 November 02 09:34 PM|