Gene therapy

A mitochondrion


Why ‘three-parent embryo’ procedure could fail

It’s a stunning and controversial procedure: Give a baby three genetic “parents” by combining sperm from dad, a cell nucleus from mom, and the egg of a female donor. The approach is supposed to eliminate the risk of inheriting sometimes deadly mutations in the DNA of the mitochondria, the cell’s energy-producing structures. But a new study corroborates what some exploring this so-called mitochondrial replacement therapy have long suspected: that the undesirable DNA still manages to sneak into the donor egg during the procedure, and mitochondria containing it could even grow to outnumber the power plants with healthy DNA.

“Researchers working in this field were aware of this potential risk,” says developmental biologist Juan Carlos Izpisua Belmonte of the Salk Institute for Biological Sciences in San Diego, California, who was not involved in the study. “But this is the first direct evidence that demonstrates that this risk is real.”

Embryos made through this mitochondrial replacement technique have only been attempted in animals thus far. But last year, the United Kingdom passed legislation approving the procedure for mothers with known mitochondrial DNA mutations. That same year, an expert panel at the U.S. National Academies of Sciences, Engineering, and Medicine endorsed clinical testing, pending a green light from the U.S. Food and Drug Administration. But some find the concept ethically troubling.

And the delicate technique isn’t perfect. Researchers must suck the microscopic DNA-containing nucleus from a mother’s cell into a tiny glass straw and transfer it to a donor egg, which has had its nucleus removed. The procedure can suck up some mitochondria, too, however, meaning that a few thousand copies of mitochondrial DNA, some potentially containing the disease-causing flaw, can get into the donor egg. And although they’re vastly outnumbered by the hundreds of thousands of copies of native mitochondrial DNA, those proportions can change unpredictably as mitochondria replicate.

“We can’t predict whether the good or the bad would win. We don’t know the rules,” says Douglas Wallace, a geneticist at the University of Pennsylvania, who has studied mitochondrial genetics for decades and was not involved in the new study. If the “bad” DNA takes over cells in the developing embryo, some tissues or organs may end up diseased.

Animal testing of mitochondrial replacement has already revealed some potential risk. Monkey embryos that developed from the procedure, for example, seemed to retain the hitchhiking mitochondrial DNA in some of their cells. But in human cells, it wasn’t known whether this DNA would disappear, or if it could make a home for itself long-term.

So in the new study, Dieter Egli, a stem cell biologist at the New York Stem Cell Foundation in New York City, and colleagues tracked proportions of carryover mitochondrial DNA, not in actual human embryos but in human stem cells created through two different types of nuclear transfer. They made eight cell lines by moving nuclei from one egg cell to another and letting the egg develop into a clump of cells known as a blastocyst. Another 12 lines came from embryos grown from human eggs that received new nuclei from somatic—i.e. nonsperm and egg—cells. (That replacement technique is being explored as a cell therapy to treat mitochondrial disease, outside of reproduction.) For these experiments, none of the mitochondrial DNA contained disease-causing mutations.The work was supported by the New York Stem Cell Foundation, as the U.S. National Institutes of Health won’t fund research that harms human embryos.

Although the large majority of the cell lines received some foreign mitochondrial DNA in the transfer, most also seemed to stamp it out, leaving no trace of it after several divisions. But three of the lines—one from the blastocyst and two from the somatic cell transfer—managed to swing the other direction: After dividing and being cultured dozens of times, they produced colonies that contained only the carryover DNA, the authors report online today in Cell Stem Cell. Cells from these lines developed into cardiac muscle and connective tissue, which also showed fluctuating levels of the foreign DNA in their mitochondria, sometimes reaching 100%.

Having all one population of mitochondria or the other didn’t seem to make a cell more competitive or healthy, Egli notes, so the competition seems to play out within cells, not between them. What gives some mitochondrial DNA copies the advantage in this microscopic battle isn’t clear. And the battle may play out very differently in a developing embryo carrying mutant DNA than it did in these dishes of repeatedly cultured cell lines with only healthy DNA, experts warn.

But if the risk is real, how could mitochondrial replacement procedures avoid it? It might not be possible to screen out these embattled cells in favor of more stable ones, the authors note, because the key mitochondrial replications occur only after the embryo is implanted in the uterus. Instead, reducing the risk will mean reducing this carryover before the battle can start, Belmonte says. His own lab is working on a way to target and destroy undesirable mitochondrial DNA in eggs or embryos using targeted DNA-cutting enzymes.

In the meantime, the possibility of failure should make researchers more cautious as they move toward human testing, Belmonte says. But he points out that for women hoping to conceive, the risk is relative. “These techniques are the only hope at this moment for families suffering from mitochondrial diseases.”