NEW YORK (GenomeWeb News) – Subsets of human skin cells from the same individual can contain slightly different copy number variant profiles in their genomes, according to a new study.
As they reported online yesterday in Nature, researchers from Yale and Stanford Universities generated multiple iPSC lines per person for seven individuals using skin fibroblast cells as the source of somatic cells for iPSC reprogramming. From genome sequence data on these reprogrammed stem cells, investigators uncovered cell line-specific deletion and duplication patterns, with two validated CNVs occurring per iPSC line, on average.
But additional analyses indicated that many of these differences were not side effects of the reprogramming process itself. Rather, roughly half of the CNVs were found at low frequency in the original skin fibroblast populations, pointing to authentic copy number mosaicism in these somatic cells.
"We found that humans are made up of a mosaic of cells with different genomes," co-corresponding author Flora Vaccarino, a neurobiology, neurodevelopment, and regeneration researcher at Yale University, said in a statement.
From their results so far, the investigators estimated that somatic CNVs exist in the genomes of nearly one-third of fibroblast cells. And they suspect that similar patterns could exist in other tissues as well — a possibility that has implications for those working to track down and assess disease-related mutations and CNVs.
"The observation of somatic mosaicism has far-reaching consequences for genetic analyses, which currently use only blood samples," Vaccarino said. "When we look at the blood DNA, it's not exactly reflecting the DNA of other tissues such as the brain. There could be mutations that we're missing."
Past studies have found that iPSCs produced from human or mouse somatic cells sometimes contain CNVs and other genetic and/or epigenetic features not found in the original cell source. But the extent to which this occurs is a matter of debate. In a study published last year, for example, researchers reported finding very few CNVs or rearrangements in mouse iPSCs relative to their original cell source.
To delve further into potential ties between cellular reprogramming and de novo CNVs, the authors of the current study used Illumina's HiSeq 2000 to do whole-genome sequencing on 20 iPSC lines — up to three lines per person — produced via reprogramming of skin fibroblast cells.
The original primary skin cell samples came from seven individuals belonging to two families. To verify that the samples were completely reprogrammed in each iPSC line, researchers relied on methods ranging from array and reverse-transcription PCR experiments to RNA sequencing and methylation mark assessments.
After identifying potential CNVs from the genome sequence data, the team searched for copy number shifts specific to each iPSC line via comparisons to the human reference genome and primary fibroblast populations.
All told, 15 of the 20 iPSC lines harbored at least one validated CNV that was absent from the other lines, with each line containing an average of two verified CNVs larger than 10,000 bases.
The team's follow-up PCR, digital droplet PCR, and breakpoint sequencing analyses indicated that a significant proportion of these CNVs had not simply sprung up de novo during reprogramming. Instead, around half of the CNVs could be traced back to low-frequency variants present in skin cells prior to reprogramming.
From their findings so far, the study's authors argued that "reprogramming does not necessarily lead to de novo CNVs in iPSCs, because most of the line-manifested CNVs reflect somatic mosaicism in the human skin."
"Although we acknowledge that some CNVs might arise during reprogramming," they said, "our data suggest that reprogramming per se does not obligatorily induce de novo mutations, as at least half of the [line manifested CNVs] preexisted in parental fibroblast cells."
The team also noted that the type of iPSC analyses described in the study — or other methods involving the clonal expansion of individual cells — could prove useful for uncovering a variety of low-frequency genetic alterations within a given tissue.