A cell, with the nucleus containing the bulk of genetic material in the centre and the mitochondria shown in green, and the method for conventional IVF, which has been adapted for the three-parent process © Eraxion (left) © ktsimage (right)
In June, Britain’s Chief Medical Officer gave a cautious go ahead for work on three-parent fertility treatments. The intention is to use genetic material from the two “biological” parents, and one other woman, to lessen the chances of the child inheriting certain disorders. The prospect of the United Kingdom becoming the first country to allow this has drawn strong opposition from religious and ethics groups. Some see it as “cross[ing] the line that will eventually lead to a eugenic designer baby market,” as David King, Director of campaign group Human Genetics Alert, put it. But some scientists also think that those in favour of the technique have downplayed the uncertainties.
Three-parent in vitro fertilisation works by combining the ova of two women to prevent children from inheriting faulty genes from the “principal” mother. The second woman, or “third parent,” donates only a small number of genes to correct potentially faulty ones, while the rest of the principal mother’s genetic material remains intact. This could prevent around 50 genetic diseases from being passed on, including muscular dystrophy and Parkinson’s, while, researchers claim, having no other impact on the child.
But this may not be the case. There is reason to think that including genetic material from a third parent would alter the child. There is the risk that parents might one day submit to a treatment that not only removes the chance of a child inheriting a certain disorder, but also changes its nature. The possibility is that, despite what has been claimed, “three-parent babies” would be precisely that, with a genetic make up derived from all three adults.
The piece of genetic material upon which the treatment depends are mitochondria. Almost all of an individual’s DNA and genes are contained in the “nuclear genome” that sits at the centre of our cells. But there are 37 genes—little more than 0.1 per cent of the total—that are located separately in mitochondria. These small units are located inside human cells, separate from the bulk of the genetic material, and are sometimes described as the powerhouses or batteries of our cells. This is due to a unique function they have: they generate adenosine triphosphate, the molecule that drives cellular metabolism. Without this molecule, cells would cease to function and they would die.
Mutations within the mitochondria’s genes can cause a wide variety of serious and life threatening diseases. A crucial property of mitochondria is that in most animals, including humans, they are inherited exclusively from the mother, meaning that all descendants of an affected woman will suffer from the same condition. Muscular dystrophy, some neurological conditions, deafness, blindness and type 1 diabetes are all associated with mutated mitochondrial genes. Roughly one in 6000 children suffer from such a condition and the three-parent IVF method is intended to prevent inheritance.
In conventional IVF treatment, two or three eggs (ova) are extracted from the woman and exposed to sperm in the laboratory. The resulting embryo is implanted in the womb of the mother. But in the three-parent process, a second woman also donates ova. The nucleus of the second mother’s egg is then removed and replaced with that of the desired “biological” mother, before exposing this newly engineered egg to sperm. In this way all the mitochondria from the intended mother are removed and replaced by (hopefully) healthy mitochondria from the “third” parent.
Although as yet untested in humans, this process has a good chance of being safe and of effectively checking the inheritance of serious conditions. In cases where the procedure fails, the most likely outcome is that the resultant embryo will simply not survive, as when conventional IVF is not successful.
But is the technique a significant genetic modification of human descendants? It is a crucial question, and one that is at the core of the European Convention on Human Rights and Biomedicine. This condones modifications to the human genome “only if the aim is not to introduce any modification in the genome of any descendants.” This includes cases in which the aim of “modification” is to remove the risk of a child developing rare hereditary conditions—and so the three-parent method seems to contravene the convention.
For this reason, mitochondria have to be treated as a special case. They have to be viewed as isolated from the nuclear genome and responsible only for specific functions, such as energy production. This is the position of those in favour of three-parent treatments, who stress that mitochondria are unrelated to genes responsible for, say, individual characteristics like eye colour and personality traits. “There is no way that this technique can influence the eye colour, hair colour or any other external characteristic or make up of the baby,” Michael Rimington, Medical Director at South East Fertility Clinic, has said.
But the view that mitochondria are isolated is incorrect. It is well established that they have too few genes to account for all of the functions they perform and so rely on working with networks of genes in the nucleus to be fully operational. This collaborative element is a reflection of mitochondria’s evolutionary history, the product of a process in the early development of life in which bacteria were captured and modified to become an effective energy production system. Over time mitochondrial genomes have shrunk to a core set of genes, the rest transferred to the nucleus to participate in other complex networks of interaction that are not directly related to energy production.
However, not only do the mitochondria depend intimately on nuclear genes for their functions, the opposite is also true: a number of critical processes once thought to be controlled solely by nuclear genes also involve mitochondrial genes. The evidence for this interplay is extensive and growing. One of the earlier studies on this subject was conducted 10 years ago at the French National Centre for Scientific Research. It found that mitochondrial DNA in mice interact with the nuclear genome to modify how the animals think. The authors concluded that “several lines of evidence indicate an association between mitochondrial DNA and the functioning of the nervous system.” Other studies have made clear the role of mitochondrial genes in developmental, neurological and other processes.
A further consideration is that changing an organism’s mitochondria could potentially lead to unforeseen health problems. A US study at Indiana University published this year involved engineering new strains of fruit fly with varying types of mitochondrial and nuclear DNA. The study showed that five of the six new strains of fly were healthy and that only one, containing a specific combination of mitochondrial and nuclear DNA, developed the anticipated reproductive and developmental problems. In other words, in this case, a defect on a mitochondrial gene alone was not sufficient to cause the condition, which required a corresponding change in a nuclear gene.
Although this study was only in the fruit fly Drosophila, a staple of laboratory experiments, it has unearthed what seems like a fundamental property of mitochondrial function: that there are complex interactions between mitochondrial and nuclear DNA and the manner in which these interactions occur can dramatically affect the animal in question. In flies the consequences were highly significant, but the gene networks of humans are known to be more numerous, sophisticated and complex than in primitive life forms. This suggests an explanation for why human diseases can vary significantly in severity between individuals who carry the same deleterious mutation in their mitochondrial genes. The interaction with the nuclear material can determine the severity of the condition.
This has two implications for the three-parent process. First, these studies imply that targeting the mitochondrial genome alone might be over simplistic, in that it ignores associated mutations in nuclear genetic material. It is true that fixing the mitochondrial genome by replacing it with one from a second mother will sometimes resolve a particular problem, but there will be other conditions for which this will not work.
Second, the fact that this treatment will involve swapping not just the gene responsible for a specific inherited disorder but all of the mitochondrial DNA could have unintended consequences. Because of the complex interactions between mitochondrial and nuclear genes, it is possible that the child conceived via this process would receive characteristics from the third parent. This contradicts arguments made by proponents of the method.
This does not mean that the idea of three-parent babies cannot be justified by the great potential benefits for the families concerned. What it does mean, however, is that mitochondrial genes must not be treated as fundamentally different from nuclear genes. They are not. They differ only in where they are located and in their being inherited solely from the mother—in all other respects they are part of the same greater genome.
The mitochondrial genome can be swapped out conveniently in a way that parts of the nuclear genome cannot, and there is a real chance that this method could help free couples wanting to have children from the burden of inherited illness. But given our knowledge of how genes interact, more research is needed before a final go ahead should be granted for three-parent babies. We need to understand fully how the mitochondria and nuclear genome interact, in order to develop ways of treating inherited diseases that target both at the same time.