The science of three-parent children

2015 has already been a significant year in the field of human medicine as February saw the UK become the first country in the world to legalise the generation of so-called 'three-parent' children. This marks a milestone for preventative genetics and embryology and offers hope to many people around the UK and beyond who would be unable to have healthy children otherwise. The votes to bring this into law were fairly comfortably won by those in favour - 382 vs 128 in the House of Commons (the lower house) and 280 vs 48 in the House of Lords (the upper house) - however there have been a number of vocal opponents to the measure. In this post I hope to explain just what the process involves, and why it is considered necessary by the majority of British MPs.

A cellular energy crisis

Mitochondria, as you may recall from a previous post, are the powerhouses of our cells. They metabolise a range of molecules derived from food at use them to generate energy in the form of another molecule, ATP. You would not last long without them - just try holding your breath for a few minutes, since anaerobic respiration is all a cell without mitochondria would be able to manage. It is not surprising, therefore, that problems with mitochondrial function can be fairly nasty. Mitochondrial diseases are a range of genetic disorders in which the proper role of the mitochondria is disrupted due to mutations in one of the genes responsible for making mitochondrial proteins. These diseases never completely knock out mitochondrial function (since an embryo with such a disease could never survive to full development) but still cause severe symptoms in sufferers. Depending on the exact mutation, these can include blindness, deafness, diabetes, muscle weakness, cardiac problems, and problems with the central nervous system. Prognoses vary from one disorder to the next, but they invariably shorten lifespan, often severely. Sufferers of Leigh's disease, for example, rarely live past 7 years of age, and spend their short lives experiencing muscle weakness, lack of control over movement (particularly of the eyes), vomiting, diarrhea, an inability to swallow, and heart problems, among others. 

What makes mitochondrial diseases unique is that they can occur without the need for mutation in the nuclear DNA, i.e. the DNA stored in the nucleus of the cell and what we typically consider our genome. This is because mitochondria have genomes of their own - a remnant of the fact that mitochondria were, once upon a time, independent bacteria that subsequently formed a symbiotic relationship with our unicellular ancestors (more on this here if you're interested). The mitochondrial genome is small, containing only 37 genes compared to the 25,000 or so in the nuclear DNA, but these genes comprise some of the most vital to mitochondrial function. In many cases of mitochondrial disease, it is mutations of one of these genes that is the problem.

An unwanted inheritance 

A quirk of the separate mitochondrial genome is that you only inherit it from your mother. During fertilisation, the lucky sperm fuses with the egg and thereby provides the remaining half of the nuclear genome needed to make a person. The sperm's mitochondria are, however, not tolerated in the maternal egg and are quickly marked for destruction. This, combined with the relative numbers of mitochondria in the two cells (only a few hundred in a sperm cell compared to up to a million in an egg) mean that essentially every mitochondrion in the developing embryo and the person it becomes will come from the mother. In a real sense, you are more genetically similar to your mother than to your father, albeit marginally.

Only maternal mitochondria survive during development. Image from the NHS.

Because of this pattern of inheritance, men who suffer from mitochondrial diseases caused by mtDNA mutations can father children without the risk of passing the disease on to their children. Women, sadly, do not have this option. A woman suffering from a mitochondrial disease will have some mutant mitochondria and some healthy mitochondria in each cell of her body. The severity of the disease depends partly on the ratio of mutant to healthy; the more healthy mitochondria the better. During egg production, primordial germ cells divide into multiple eggs and the mitochondria become divided as well. Some eggs will have a better mutant:healthy ratio than other, and so the severity of the disease in a child will vary depending on which egg it is that gets fertilised.

The severity of mitochondrial disease depends on the proportion of mutant
mitochondria in the fertilised egg (oocyte). Image from the NHS.

Averting the problem

Medical science has so far found few effective treatments for mitochondrial diseases, however there is hope for female suffers who want to have children without passing on their condition. Since mitochondria are the problem, if they can be replaced by healthy mitochondria but keep the same nuclear DNA, then the child would still be genetically hers but without the disease. This is the principle behind 'three-parent' babies. In this process, the nucleus is extracted from one of the mother's eggs and implanted into the egg of a healthy donor with its nucleus removed, essentially making a healthy egg with the mother's own genome. This can then be fertilised in vitro by the father's sperm and implanted into the mother's womb for gestation, as with typical IVF. The resulting child will have DNA from three people; nuclear DNA from its mother and father, and the mtDNA of the healthy egg donor. This is what makes them a 'three-parent' child, but in reality the amount of DNA from the donor is minuscule compared to that of the 'true' parents - less than 0.1%. It has been suggested that the term 'two-and-one-one-thousandth-parent child' would be more genetically apt!

This process of mitochondrial transfer is relatively new but the concept of children bearing genetic material from three individuals is not - in fact there are a number of people alive now who have just that. This is because a similar process of cytoplasmic transfer was already in use as a fertility treatment in the US before being effectively banned in 2001. This involved the movement of healthy cytoplasm (mitochondria and all) into the host egg, rather then movement of the nucleus, but it amounts to a similar outcome. 

Why only the UK? 

Unsurprisingly this kind of process has been highly controversial in many countries and continues to stir debate even in the UK. Some opponents simply believe that any interference with the 'natural' process of fertilisation is inherently unacceptable, though arguments such as these are usually based on religious doctrine more than genuine scientific concern. Many people, however, believe that the outcomes of generating what amounts to genetically modified humans are simply too unpredictable to risk. This concern is understandable and should certainly be subject to rigorous discussion (as, indeed, it has been in both houses of parliament) however it is not one that I share. The argument boils down to the fact that mitochondria are not identical (even healthy ones), and also that mitochondria communicate heavily with the nuclear genome. Therefore, mixing the 'wrong' mitochondria with the 'wrong' nuclear DNA could have unforeseen consequences in either the resultant child or their future offspring. I can see where this argument is coming from, however I don't see why mixing mitochondrial and nuclear DNA through mitochondrial transfer is any different through doing it via normal fertilisation. Every time a child is conceived, half of their nuclear DNA is from an individual with different mitochondria to them, their father. Potentially, they may have very different mitochondria to them if, for example, the parents are from very different ethnic backgrounds. This is clearly not a hindrance to healthy development or genetic health down the generations, and I don't see a clear difference with mitochondrial transfer. Indeed, I would say the more generally accepted process of surrogacy is likely to have potentially greater impacts on genomic behaviour. This is because epigenetic processes link environmental factors to genome activity (more on this in a previous post here if you're interested) and this is particularly significant during gestation. The specific environment of a surrogate mother's womb (which may differ significantly from that of the biological mother for many reasons) will influence the development of an embryo in ways that may be passed on to that child's offspring in equally unpredictable ways as with mitochondrial transfer, yet surrogacy is widely legal across the world. 

Much of the fear surrounding procedures such as these seems to stem from the concept that there exists a single, ideal human genome from which we can only stray at our peril. Biology is much less perfect than that - genes are mixed liberally and randomly all the time in all species, and are mutated in a similarly chaotic fashion. This is not to say that we should be slapdash with any attempts we make to direct our own genetics, but we should bear in mind that there is no such thing as the 'human' genome as every individual is different and no one person is more genetically important than another. I'm hopeful for mitochondrial transfer therapy and am proud that it will the UK leading the way in relieving the suffering of the many people living with mitochondrial diseases. 

How many parents could a child have?

Finally, thinking about this topic got me wondering about the question of how many parents a child could, in theory, have? If we say that someone has to contribute a self-contained active piece of genetic material (i.e. not just one or a small number of base-pairs), then we could say that each of the 25,000 or so genes in the genome could be provided by different parents. But why stop there? There are several thousand non-coding RNAs encoded by the human genome that have all sorts of important functions, each could be donated by a different parent. Moreover, genes are regulated by a wide range of regulatory elements that are encoded within the genome but vary from person to person. It's hard to put a precise number on how many of these we have, but it's probably somewhere in the region of half a million. Given that everyone has two copies of each gene, ncRNA, regulatory element etc., a reasonable estimate of the number of parents a child could have where every parent contributed direct genetic activity is probably in the region of 1,200,000 or so. If you include the parts of the genome that don't seem to do much then that number rises to tens or hundreds of millions, but 1,200,000 seems plenty to me!

Of course, to achieve this the child's genome would have to be synthesised to contain the matching sequence from each parent - the DNA molecules within the first cell of the embryo would never have physically been inside the parents, unlike normal fertilisation. Some people might therefore say that they don't count as parents at all, so the child could either be considered to have over a million parents or none at all. With this in mind, three parents doesn't seem like too much of a leap after all!