NEW YORK (GenomeWeb) – With the help of a new single-cell analysis approach, researchers have uncovered thousands of copy number variants that occur in developing brains.
Brain cells accumulate somatic changes — often copy number changes — that lead to genomic mosaicism and cellular diversity. But how these neural CNVs arise has been unclear.
Researchers from the Sanford Burnham Prebys Medical Discovery Institute in La Jolla, California, and elsewhere developed a single-cell analysis approach that relies on machine learning to detect small copy number changes. As they reported this week in the Proceedings of the National Academy of Sciences, they applied this approach to cells from developing mouse brains to uncover thousands of alterations.
"This study fills critical holes in our understanding of copy number variations in the brain and provides important clues for further study," senior author Jerold Chun, a professor and senior vice president of neuroscience drug discovery at SBP, said in a statement. "We show that a great number of CNVs in single brain cells arise before birth as the brain begins to form and are later incorporated into the mature brain, indicating they are foundational to the brain's cellular diversity and development."
The researchers applied this approach to samples obtained from 43 mice throughout cerebral cortical neurogenesis. They also collected cortical neurons and splenocytes from adult mice as controls. In all, they amplified DNA from 658 single cells, with 488 cells passing quality control.
Chun and his colleagues relied on a transposase-based amplification (TbA) approach that combines whole-genome amplification and sequencing library preparation to generate a large enough amount of DNA for sequencing. When they compared TbA to Sigma's GenomePlex amplification kit, the researchers found that TbA exhibited lower noise and allowed for the identification of both more and smaller CNVs.
They coupled that amplification approach to a bioinformatics method, called filtering unreliable CNVs (FUnC), to weed out suspect CNV calls. They trained this approach using lymphocyte V(D)J recombination — immune cells in which DNA is recombined in a reproducible way. They reported that TbA-FunC provided more genomic coverage and removed more than 90 percent of false-positive CNV calls.
The investigators reported that they uncovered thousands of new CNVs, about half of which were smaller than 1 megabase. More than 93 percent of the samples harbored one or more CNVs, they added. In addition, these cortical CNVs appeared to be scattered throughout the genome, in contrast to V(D)J recombination.
While the researchers noted both deletions and duplications, they found deletions to be more common. In particular, this led to cumulative DNA loss between an early and later embryonic stage.
The researchers also found that the number of CNVs increased until mid-neurogenesis. It peaked at double the rate seen in adult neurons and at three times that seen earlier in fetal development. This coincides with the peak of programmed cell death and suggests a relationship between the two processes, they said. There could be a threshold of CNV production that leads to cell death, they added, and it also suggests that CNV generation is a regulated process.
"These findings demonstrate that the fetal brain is patterned by a complex mosaic of myriad CNVs before birth, diversifying the individual cells that make up our brains," Chun said. "Our brain's tabula rasa may receive its first writing, which could remain with us for life, through this process."