What’s the best recipe for creating patient-specific stem cells? The question has prompted years of debate and a series of contradictory papers. Ideally, the cells should develop into any cell type just as well as those in a natural embryo do. For now, the 7-year-old method of introducing a small set of genes into adult cells to create so-called induced pluripotent stem (iPS) cells dominates the field. But a few researchers have stuck with an older approach—putting a donor nucleus into an unfertilized egg and deriving stem cells from the resulting embryo. This somatic cell nuclear transfer (SCNT) technique is costly, technically challenging, and ethically fraught on several fronts. But proponents have argued that cells created this way are superior, in part because iPS cells may retain features of mature cells and could acquire a greater number of genetic changes in the reprogramming process. The latest research on the matter, published this week in Cell Stem Cell, examines genetic features of the two cell types and declares a tie.

That might surprise some people, because the researcher who led the work, Dieter Egli of the New York Stem Cell Foundation (NYSCF) in New York City, has long been a proponent of nuclear transfer. He gave a preview of the paper on 22 October at the foundation’s annual conference. “This means that all of you who are working on iPS cells are probably working with cells that are actually very good. So I have good news for you,” he told them, eliciting murmurs and chuckles. “What this exactly means for the SCNT program, I don’t know yet.” Indeed, some longtime champions of the nuclear transfer—Egli included—aren’t ready to abandon the method yet.

The 2006 discovery by Shinya Yamanaka and colleagues that inserting just a few genes into mature cells could make them behave like immature ones revolutionized the stem cell field. Labs have since refined the technique to create patient-specific cells for studying disease and have already begun clinical testing of cells derived from iPS cells to restore damaged or diseased tissue.

But some wondered if this genetic approach to reprogramming created undesirable changes—even cancer-causing mutations. “We were very worried,” says Nizar Batada, a cancer biologist at the Ontario Institute for Cancer Research in Toronto, Canada. He co-authored a 2011 Nature paper showing that iPS cells had more genetic abnormalities than the skin cells used to derive them. “You have a normal cell sitting around, expecting to be a skin cell, and you’re blasting it with genes, so there’s chaotic change in DNA replication and gene expression.” Other studies of DNA methylation—the addition of chemical tags that control gene expression at certain locations—hinted that iPS cells bore some memory of their past, as adult skin cells, for example, and could be less efficient at differentiating into new cell types.  

Some scientists believed that nuclear transfer—the method used to clone Dolly the sheep in 1996—creates stem cells that are less susceptible to these issues. But there was a catch: By the time Yamanaka shared the 2012 Nobel Prize in physiology or medicine with John Gurdon for nuclear transfer, no one had yet used SCNT to create human stem cells, and the field’s attention had largely shifted to iPS cells. Last year, a team led by Shoukhrat Mitalipov at Oregon Health & Science University in Beaverton finally declared success with SCNT in human cells, and the debate was revived.

Egli and colleagues wanted to know whether the newly available cells really held advantages over iPS cells. So far, most studies comparing the two methods have had a common shortcoming, Egli says: They start with cells derived from different donors. Possible genetic differences between individuals make it hard to tell how the reprogramming techniques themselves altered the nuclear DNA.

Egli and colleagues instead used skin cells from two people—a newborn and an adult—to create both SCNT-derived stem cells (using donor eggs) and iPS cells. Then they compared the two types of cell lines with the original skin cells in terms of genetic mutations, changes in gene expression, and differences in DNA methylation. Both methods resulted in about 10 mutations compared with the average genome of the mature source cells. These changes didn’t necessarily happen during reprogramming, however, Egli says. Many were likely present in the original skin cells, and some could have arisen during the handling of cells before they were reprogrammed.

Both types of stem cells also carried a similar amount of methylation changes. Overall, the method didn’t seem to matter, Egli and his team concluded. Because he is a longtime proponent of SCNT, Egli says it would have been “more attractive” to reveal significant differences between the two kinds of stem cells. “This is simply not what we found.”  

The study doesn’t say anything about whether the changes documented in both cell types make them less fit for use in research or therapy, Batada notes, or even if the mutations observed affect the genes’ activity. “Whether these aberrations have any function isn’t clear,” he says, but in terms of the two stem cell creation methods, “they’re equally good, or equally bad.”

Mitalipov, who also contributed to the Cell Stem Cell paper, maintains that nuclear transfer creates more stable cells that more closely resemble embryonic cells. He published research this summer reporting that the methylation patterns in nuclear transfer cells more closely mimic natural embryonic stem cells, and has work in progress to identify other advantages.

Juan Carlos Izpisua Belmonte, a developmental biologist at the Salk Institute for Biological Studies in San Diego, California, finds the lack of differences between the two stem cell types surprising. But this study is “probably not enough to settle this issue,” he says. “There are and will always be concerns related to the safety [of stem cells] regardless of the methodology,” he adds, and there’s “no reason so far to discredit one in favor of the other one.”

Both Egli and Mitalipov plan to continue their work on nuclear transfer and defend its value. Egli points out that the research led to other important discoveries, including technology that  may allow a woman to avoid passing on mitochondrial diseases to her children using an egg donor. And he suggests that therapies based on donor eggs may have an easier path to regulatory approval. Until either type of stem cells results in an approved therapy, Egli says, “I think it would be a big mistake to put all eggs into just one basket.”