In September, reproductive endocrinologist John Zhang and his team at the New Hope Fertility Center in New York City captured the world's attention when they announced the birth of a child to a mother carrying a fatal genetic defect.
Using a technique called mitochondrial replacement therapy, the researchers combined DNA from two women and one man to bypass the defect and produce a healthy baby boy — one with, quite literally, three genetic parents.
It was heralded as a stunning technological leap for in vitro fertilization, albeit one that the team was forced to perform in Mexico, because the technique has not been approved in the United States.
The technique is spreading quickly, gaining official approval this month from the Human Fertilization and Embryology Authority in the U.K. The move will allow clinics to apply for permission there to carry out the treatment, with the first patients expected to be seen as early as next year.
But for all the accolades, the method also has scientists concerned that the fatally flawed mitochondria can resurface to threaten a child's health.
Earlier this month, a study published in Nature by Shoukhrat Mitalipov, head of the Center for Embryonic Cell and Gene Therapy at the Oregon Health and Science University in Portland, suggested that in roughly 15 percent of cases, the mitochondrial replacement could fail and allow fatal defects to return, or even increase a child's vulnerability to new ailments.
The findings confirmed the suspicions of many researchers, and the conclusions drawn by Mitalipov and his team were unequivocal: The potential for conflicts between transplanted and original mitochondrial genomes is real, and more sophisticated matching of donor and recipient eggs — pairing mothers whose mitochondria share genetic similarities, for example — is needed to avoid potential tragedies.
"This study shows the potential as well as the risks of gene therapy in the germline," Mitalipov says. This is especially true of mitochondria, because its genomes are so different than the genomes in the nucleus of cells. Slight variations between mitochondrial genomes, he adds, "turn out to matter a great deal."
Mitochondria are the energy powerhouses inside our cells, and they carry their own DNA, separate from our nuclear genome.
The danger lies in the fact that mitochondria are in some ways like aliens inside our cells. Two billion years ago they were free-floating bacteria basking in the primordial soup. Then one such microbe merged with another free-floating bacterium, and over evolutionary time, the two formed a complete cell. The bacteria eventually evolved into mitochondria, migrating most of their genes to the cell nucleus and keeping just a few dozen, largely to help them produce energy.
Today, our nuclear genome contains around 20,000 genes, while a scant 37 genes reside in the mitochondria. And yet the two genomes are intensely symbiotic: 99 percent of the proteins that mitochondria import are actually made in the nucleus.
Mitochondria also still divide and replicate like the bacteria they once were, and that constant replication means that mutations arise 10 to 30 times more often in mitochondrial genes than in the nucleus. If too many mitochondria become dysfunctional, the entire cell suffers and serious health problems can result. Faulty mitochondria are implicated in genetic diseases, as well as many chronic conditions from infertility to cancer, cardiac disease and neurodegenerative diseases. That's because when mitochondria falter, the energy system of the cell itself is compromised.
A three-parent baby could solve the problem by overriding faulty mitochondria, but it also raises the stakes, because the procedure does not completely replace the defective mitochondria with healthy ones.
When the mother's nucleus is transferred, it's like a plant dug up out of ground — a bit of the original soil (in this case, the mother's mitochondria) is still clinging to the roots. That creates a situation that never happens in nature: Two different mitochondrial genomes from two different women are forced to live inside the same cell. In most cases, a tiny percentage (usually less than 2 percent) of the diseased mitochondria remain — but that tiny percentage can really matter.
In his new study, Mitalipov crafted three-parent embryos from the eggs of three mothers carrying mutant mitochondrial DNA and from the eggs of 11 healthy women. The embryos were then tweaked to become embryonic stem cells that could live forever, so they could be multiplied and studied. In three cases, the original maternal mitochondrial DNA returned.
"That original, maternal mitochondrial DNA took over," Mitalipov says, "and it was pretty drastic. There was less than 1 percent of the original maternal mitochondrial DNA present after replacement with donor DNA and before fertilization, and yet it took over the whole cell later."
Mitalipov warns that this reversal might not only occur in the embryonic stem cells; it could also occur in the womb at some point during the development of a baby. Complicating things further, Mitalipov found that some mitochondrial DNA stimulates cells to divide more rapidly, which would mean that a cells containing the maternal mitochondrial DNA could eventually dominate as the embryo developed.
Some mitochondrial genomes replicate much faster than others, says University of California molecular biologist Patrick O'Farrell, who called Mitalipov's research both impressive and in keeping with his own thinking on the matter.
A diseased mitochondrial genome could behave like a super-replicating bully, O'Farrell says, re-emerging and having a large impact on the three-parent baby at any time. It could also affect that child's future offspring. "The diseased genome might stage a sneak comeback to afflict subsequent generations," O'Farrell says. On the other hand, he says, the super-replicators could act as "superheroes," if they carry healthy, fit DNA that is able to out-compete a mutant genome.
The nuclear genes donated by a father could also influence the behavior of the mitochondria in ways we cannot yet predict, O'Farrell says. For example, the father might introduce new genes that favor the replication rate of a defective bully genome. Or the father might introduce genes that help a "wimpy" healthy genome survive and thrive.
Mitalipov's proposed solution to the problem is to match the mitochondria of the mother and the donor, since not all mitochondria are alike. Human mitochondria all over the earth are in a sense a billion or more clones of their original mother, passed down in endless biblical begats from mother to child. Yet, even as clones, they have diverged over time into lineages with different characteristics. These are called haplotypes.
O'Farrell mentions blood types as a comparison. Just as you would not want to transfuse blood type A into someone with blood type B, you might not want to mix different lineages. And while he says he thinks the idea of matching lineages is brilliant, he suggests going a step further. "I say let's ... try to get a match with the dominating genome so that the defective genome will ultimately be completely displaced."
In fact, he adds, the ideal would be to look for one superhero genome, the fastest replicator of all – one that could displace any diseased genome.
To find out which branches are super replicators, O'Farrell hopes to collaborate with other laboratories and test the competitive strength of different haplotypes. Earlier this year, O'Farrell's laboratory published work showing that competition between closely related genomes tends to favor the most beneficial, while matchups between distantly related genomes favor super replicators with negative or even lethal consequences. There are, he says, at least 10 major lineages that would be distinct enough to be highly relevant.
Mitalipov says that most of the time, matching haplotypes should ensure successful mitochondrial transfers. However, he cautions that even then, tiny differences in the region of the mitochondrial genome that controls replication speed could cause an unexpected surprise. Even in mitochondria from the same haplotype, there could be a single change in a gene that could cause a conflict, he says.
In his study, Mitalipov zeroes in on the region that appears responsible for replication speed. In order to find out a mother's haplotype, he says, full sequencing is necessary, and this region from the donor's egg should also be looked at to be sure it matches the mother's. Today, it costs a few hundred dollars to sequence a woman's mitochondrial genome.
But battles between mitochondrial genomes are only one part of the emerging story. Some research suggests that nuclear genes evolve to sync well with a mitochondrial haplotype, and that when the pairing is suddenly switched, health might be compromised.
Research in fruit flies and in tiny sea creatures called cephalopods shows that when the "mitonuclear" partnership diverges too much, infertility and poor health can result. In some cases, however, the divergent pairs are above average and can actually lead to better health.
Swapping as little as 0.2 percent of mitochondrial DNA in laboratory animals "can have profound effects on the function of cells, organs, and even the whole organism, and these effects manifest late in life," according to mitochondrial biologist Patrick Chinnery of the University of Cambridge, writing in November in The New England Journal of Medicine.
Because of all these unknowns, a U.S. panel recommended last February that mitochondrial replacement therapy, if approved, implant only male embryos so that the human-altered mitochondrial germline would not be passed down through the generations.
Most scientists approve of this advice, but biologist Damian Dowling of Monash University in Melbourne, Australia, has reservations about even this solution.
His own research in fruit flies shows that males may actually be more vulnerable than females to impaired health from mitochondrial replacement. Since females pass on mitochondria, natural selection will help daughters sift out any mutations that might be harmful to them, and keep their nuclear and mitochondrial genes well matched. Males aren't so lucky: If mutations don't harm females but do harm males, the males may have to suffer impaired fertility and go to their graves earlier.
This is known as the "mother's curse" — a term coined by geneticist Neil Gemmell of the University of Otago in New Zealand to describe the biological baggage that mothers unwittingly pass down to their male babies.
The bottom line, according to biologist David Rand of Brown University, who studies mitochondrial genomes, is that when you swap mitochondria, the reaction is "highly unpredictable."
And that's why many experts are calling for caution even amid all the excitement following the three-parent Mexico trial — though there is reason to believe they aren't being heard.
A three-person baby has now been born in China, and two more may soon be born in Ukraine, according to Nature News. Zhang, meanwhile, continues to encourage potential patients in Mexico: "We have received interest both locally and abroad," he says, "and we invite people to learn more about the treatment."
Doug Wallace, head of the Center for Mitochondrial and Epigenomic Medicine at the Children's Hospital of Philadelphia, is among those calling for a more methodical approach to the technique, though he says he doesn't think there's any way to put the brakes on now. "I think what's happened is we're going to see more and more trials and some families are going to be exceedingly fortunate — and perhaps some will be an unfortunate part of the learning set."
Research on mitochondria has to catch up, Wallace says, and while matching haplotypes is a good idea, it isn't so easy to do in practice. "Finding women to be egg donors is going to be a major limitation," he says — especially when you'd first have to survey a large group to find compatible mitochondrial DNA.
Still, for women desperate to conceive a healthy child this may seem reasonable. Wallace adds that mitochondrial replacement therapy might find favor even outside those seeking to avoid passing on fatal genetic mutations — such as older women simply facing reduced fertility. "There's no proof that's the case," he says, but if it came to pass, that could mean a therapy that might change the DNA of tens of thousands, maybe hundreds of thousands, of babies conceived by this method.
That would have a real impact on the long-term future of society, Wallace adds, and we don't yet fully understand all of the implications.
"I think it's an exciting possibility," he says, "but also a little disconcerting."
Jill Neimark is an award-winning science journalist and an author of adult and children's books. Her most recent book is "The Hugging Tree: A Story About Resilience."
A version of this article originally appeared at Undark, a digital science magazine published by the Knight Science Journalism Fellowship Program at MIT.