Although the many cells in a human body have distinct functions and appearances, it’s generally been assumed that they all share the same genetic blueprint. So when adult cells are reprogrammed into their most basic, stem cell state, it’s assumed that the resulting stem cells will all be the same. Such induced pluripotent stem cells (iPSCs), the thinking goes, could then be coaxed to develop into one of a number of different cell types that genetically match a donor. But a new discovery by a team of Yale researchers has upended this reasoning: cells accumulate so many genetic changes during a human’s lifetime, they’ve found, that even a single tissue can give rise to genetically diverse iPSCs.

“These cells are increasingly used as models for disease and potentially can be used as the basis for treatments,” says Flora M. Vaccarino, M.D., Harris Professor in the Child Study Center and professor of neurobiology, who led the new study. “But there was evidence based on other experiments that there was genetic variation among populations of iPSCs, which could be bad news for the field.”

The variation had been spotted when other researchers compared the genomes of iPSCs that they expected to be identical, since they’d all been reprogrammed from the same tissue in a single individual. Instead, when the iPSC genomes were compared to one another, huge chunks of DNA were found to be duplicated or deleted—a phenomenon called copy-number variation (CNV). Scientists began to fear that reprogramming creates unstable genomes and an increased ability to develop mutations, which would undermine the promise of iPSCs for both research and therapy.

But Vaccarino and her collaborators, including first author Alexej Abyzov, Ph.D., associate research scientist, and co-senior authors Mark B. Gerstein, Ph.D., the Albert L. Williams Professor of Biomedical Informatics, and Alexander Urban, Ph.D., of Stanford University School of Medicine, thought more work was needed to show exactly what was causing these CNVs, so they launched a detailed genomic study of a group of iPSC cell lines that originated from skin cells of seven individuals. Whole-genome DNA sequences were obtained from three iPSC lines from each donor and were compared through several bioinformatic approaches to that of the donor’s skin cells.

“The first thing we found was that there was, in fact, an alarming number of copy number variations, both duplications and deletions,” says Vaccarino. Each iPSC line had an average of two CNVs, she says, though some had as many as five. “But we were still not convinced that this was due to the reprogramming.” Then they were surprised to notice that two different iPSC cell lines that originated from the same person had an identical CNV. For both lines to have developed precisely the same variation independently was highly unlikely, and the observation suggested that the variation already existed in the donor’s skin cells and was not due to the reprogramming used to make the iPSCs.

So Vaccarino and her collaborators—a multidisciplinary team that included stem cell biologists, bioinformaticians, and geneticists—turned to a new, high-resolution technology called digital PCR to scour the original skin cells for CNVs. Unlike older technologies, digital PCR is sensitive enough to detect variations present in only 0.1 percent of the cells. The team discovered that around half of the CNVs they’d pinpointed in the iPSC lines could also be found in the fibroblasts. “It could be that even more than half are present,” says Vaccarino, “but that’s what we were able to detect with this method.”

The study’s most intriguing twist is that the researchers found iPSCs to be remarkably stable—reprogramming does not appear to significantly alter the genomes of donor cells. Instead, it is donor cells that show an unexpected amount of variation.

Genomic variation in the cells of a single individual is called mosaicism, after the differently colored tiles that make up a mosaic. Mosaicism is known to result from the cell dysregulation seen in cancer and other diseases. But significant genetic differences among cells in healthy individuals were thought to be rare, especially among cells in a single tissue.

The new findings, which were published in the December 20, 2012, issue of the journal Nature, suggest that such variation has been “markedly underestimated,” write the authors.

“In the skin, this mosaicism is extensive and at least 30 percent of skin cells harbor different deletion or duplication of DNA, each found in a small percentage of cells,” Vaccarino explains. “This has far-reaching consequences for genetic analyses, which currently use only blood samples. 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.”

The good news for researchers moving forward, says Vaccarino, is that iPSC cell lines provide a straightforward way to reveal genetic diversity within individuals. “We can now use the stem cells as a discovery tool to look at these rare events.” she says.