8

How to make a baby

All you need is love

JOHN LENNON

There’s a slogan popular among those opposed to same-sex marriage: ‘God made Adam and Eve, not Adam and Steve’. Increasingly, around the world, Adam and Steve can get married. They can even have children — but only with the help of someone to provide an egg, and someone to carry the baby. We’re nowhere close to developing an artificial womb, but could Adam and Steve dispense with the egg donor, and have children to whom both are biological parents? Equally, could Anna and Eve skip the sperm donor?

The answer to that question may soon be yes.

The question might also be posed this way: could Adam make an egg? Could Eve make a sperm?

Which leads to the question: what are eggs and sperm, anyway? And could we make them without the aid of ovaries and testicles? Surprisingly, the answer lies in part with a little boy, hovering on the edge of death.

James was just three when, one day, all of his muscle cells fell apart.

He had been a happy, healthy boy, always on the go, an explorer. Like any small child, he had had his share of colds and other minor illnesses, but nothing out of the ordinary. Then he ran into a virus that stressed his body just slightly more than the others had — and uncovered a vulnerability that had always been there, but had never before revealed itself. James woke one morning cranky and irritable, complaining of pains in his legs. Then he collapsed. His mother called an ambulance, which only just got him to the hospital in time.

Your muscles contain an enzyme called creatine kinase, shortened to CK. It’s an important enzyme, but for this story it doesn’t really matter what it does — just that muscle cells contain a lot of it, and, when they die, CK is released into the bloodstream. Muscle cells wear out and are replaced all the time, so there’s always a little bit of CK in your blood. Normally, that would amount to less than 200 units of enzyme per litre — more if you’ve just run a marathon, but not usually a lot more. The first time I heard about James was from the intensive care specialist who was looking after him. He was marvelling at James’s CK. ‘500,000! It must be a world record!’ And he thought the problem must surely be genetic, which was why I was there.

Record or not, it was bad news for James.

The problem did turn out to be genetic — James had two faulty copies of a gene called LPIN1, which is important for keeping the lining of muscle cells stable and strong. In its absence, his muscle cells were fragile, needing only a small push — for example, from what would otherwise be a mild viral infection — to make them break down completely. The same virus that might have given you a runny nose and some aches and pains was potentially a deadly threat to James.

When cells die, other substances pour into the blood. Sometimes, the effects of this can be fatal; cells are rich in potassium, but too much of it in the bloodstream can stop your heart. James had escaped that fate, but he was still in terrible trouble. His muscles felt hard to the touch, like wood wrapped in rubber. We worried that he might have a problem called compartment syndrome, in which swollen muscles are trapped inside their sheaths and the resulting rise in pressure blocks off their own blood supply, so that the muscle dies completely. The surgeons decided to do an operation to check for this, and if necessary release the pressure in his leg muscles. After a single cut, they closed him up again, sure there was nothing to be done. The muscle looked ghastly — pale, bloodless, and seemingly beyond hope.

Without muscles, you can’t walk or use your arms and hands, and you can’t breathe. James was totally paralysed, and most of us thought he was going to stay that way. Only one of the team looking after James was optimistic about his chances. His neurologist, a man of long experience in the field, told us, and his parents, that he thought there was every reason to be optimistic, and that James could make a full recovery. I found it hard to believe, and felt cross with my senior colleague for giving the parents false reassurance.

Six months later, James walked into my consulting room looking as though he had never been sick.

How was this possible? Although many cells in each of his muscles had perished, some had survived. And among the survivors were some that were special.

They were stem cells.

Sperm are relatively simple critters. They are little guided missiles, carrying only the bare essentials needed to get to their targets. In the head of a sperm is the payload — half of the man’s genome (he’s only potentially ‘the father’ at this point). There’s a mid-section full of mitochondria — that’s the power plant. And there’s the tail, a thrashing motor driven by that power plant. A sperm has one job, and one job only: get the payload to its target. When a man ejaculates, there are generally hundreds of millions of sperm released. Most of the time, there is no egg waiting for them, and they die, unmourned and unfulfilled. But even if there is an egg, only one of all those millions can fertilise it. A man might produce 500 billion sperm in his lifetime, with only one or two fulfilling its destiny. If you were naturally conceived, the sperm that makes up half your DNA was the winner of a contest against incredible odds. The same goes for each of your parents, and their parents, and so on. If you ever find yourself lying awake at night wondering if there is anything special about you, consider this — you are the culmination of a long, long line of outstandingly lucky sperm, stretching back over many millions of years. They were all great swimmers, to be sure — but mostly they were freakishly fortunate. There are many, many ways for a sperm to fail to reach or fertilise an egg. The odds against your existence are genuinely astronomical. You are, without a shadow of a doubt, something special.

Once that magical, one-in-hundreds-of-billions of sperm hits its mark, its job is done. It is swallowed up by the egg, which gathers up the DNA in the head of the sperm and hunts down and kills the DNA in the sperm’s mitochondria — and then the real work can begin.

To understand how remarkable an egg is, you first need to know the difference between a muscle cell and a liver cell. Why is one of them good at contracting powerfully, while the other would never dream of doing such a thing, but excels at cleaning toxins from your bloodstream, and manufacturing the proteins your blood needs so it can clot when you’re bleeding? Both cells have exactly the same DNA — but they use it in different ways.

You can think of the genome as a box full of electrical components, all wired to a single giant circuit board. There’s everything you need to make a television, everything you need to make a hairdryer, everything you need to make a microwave oven, and so on. It’s said that there are over 200 types of cell in the human body. There’s reason to think there might be considerably more than that — there may be important differences between a skin cell on your elbow and one on the tip of your nose, for example. But let’s say there are only 200. This means that the box of components in the cell nucleus has everything needed to make any one of 200 different electrical items. There are some things that almost every type of electrical equipment needs — a way of getting electricity from the outlet, for instance — and there’s only one of those in the box. Similarly, there are some components that every cell needs, such as the toolkit that is used for disposing of damaged proteins. On the other hand, only your fridge needs a compressor, just as only a particular set of cells in your pancreas needs to make insulin.

Let’s say our electrical box is set up so there are switches that select whether any one component is used or not, and let’s pretend it doesn’t matter in what order they are joined up. This means that, by choosing which of the components in the box to switch on, you can make it behave very differently. The same box could function as a computer, or a printer, or an electric mixer, or a bandsaw. That’s pretty much what happens in a cell, too. Each cell has exactly the same set of 23,000 genes, but in each type of cell only a particular set of those genes is switched on — the rest are silenced, more or less permanently. There’s a set of genes that are switched on in every cell, known as housekeeping genes; there are genes that are needed by multiple cell types, but not all; and there are some genes, like the insulin gene, that are specific to just one type of cell.

The special thing about the egg is its potential. A fertilised egg can — and must — give rise to every other type of cell, including the eventual eggs or sperm that will become the next generation. I like to imagine this first cell as vibrating with energy, bursting with potential. That first cell, and all of its daughter cells for the first few cell divisions, are the ultimate stem cells. They are totipotent — literally, ‘wholly powerful’ — meaning that each of those cells is uncommitted, and can become any of the hundreds, or perhaps thousands, of cell types needed to make a person — and also the placenta, needed to support and nourish the baby until it is ready to be born. One step down are the pluripotent stem cells, which are ‘severally powerful’ — although the only thing they can’t make is a placenta. Identical twins are evidence of the power of pluripotent stem cells — if something causes an early embryo to split, you get two babies for the price of one.

As the embryo develops, however, cells start to set off down particular paths, gradually becoming more and more committed until most of them have reached their final form (a bit like an evolving Pokémon). Once a liver cell is a liver cell, that’s all it will ever be. The molecular switches that choose the particular set of genes needed by that cell type are welded into place. The television will only ever be a television; the blender will never download an ebook.

Except … except for the cells that don’t go all the way down that path. In every part of your body, there are cells that didn’t quite commit all the way. Most of them sit around waiting to repair damage — these are the ones that saved James, by regrowing his muscles.80 Some are very active, like the ones that live in your bone marrow making new blood cells, or the ones that replenish the cells in your gut, which are always being worn away by the contents of your bowel. Others seem to sit around for a very long time, ready for when they are needed.

[1 The stem cells in muscle are called satellite cells. When muscle is damaged, they divide; some of the daughter cells remain as satellite cells, ready if they are needed in the future. The rest merge with the damaged mature muscle cells and repair them.]

Stem cells can be more or less specialised, too. A haemocytoblast can become any type of blood cell. Once it commits to becoming a megakaryoblast, it’s still a stem cell, but all it can ever make is megakaryocytes. These, in turn, are the weird cells that make platelets, the little cellular scraps that circulate in your blood, waiting for the chance to help make a clot. Megakaryocytes — unlike the tiny platelets that they make — are huge, and have an enormous nucleus, with extra sets of chromosomes. While they are forming, they double and redouble their chromosomes, and they can have as many as 32 times as many chromosomes as a cell usually does. We have no idea why they do this, by the way.

So — stem cells are just cells that have the flexibility to become other types of cell. Unfortunately, the power of stem cells is limited. Often, damage to tissue leads to a scar, rather than replenishment by stem cells. Still, there is a lot of interest in using stem cells as medical treatments, getting them to repair damaged tissue — such as after a heart attack — with healthy new cells. For our purposes, though, what matters is that stem cells have this flexibility, and that the fertilised egg is the ultimate stem cell, ready and able to become every other type of cell.

For Adam and Steve’s purposes, we don’t need Adam to make a fertilised egg — we can get sperm from Steve, after all. But we do need to be able to persuade a cell that has gone down one path in life to change its mind and mature first into the precursors of an egg and then into an actual egg. Making an egg is a complex process, but, if we could persuade any cell type to become another cell type, there’s no reason the second cell type couldn’t be an egg.

Can we do that persuading? In principle, there’s no reason why not.

Professor Richard Harvey, who was my PhD supervisor, is an eminent biologist who studies the way the heart develops. Richard’s lab uses a variety of different methods to try to understand the complex processes that lead to the formation of the heart and all its structures. Once, I was visiting the lab and Richard beckoned me over to a microscope and asked if I wanted to see something cool. There’s only one answer to that question … and he delivered, in spades. Looking through the microscope, I saw a glass slide with clumps of cells growing on it. The cells were pulsating rhythmically. These were cardiac organoids: clumps of cells that had been persuaded that they were actually a heart, and were behaving as though this were true, pumping away faithfully. They had been made by treating stem cells with cellular messages that said: you will be part of a heart.

Even cooler, Richard’s lab had started out by making the stem cells from mature skin cells. In other words, it is already possible to make stem cells by starting with mature, fully committed adult cells and hitting reset. What you get are known as induced pluripotent stem cells — iPSCs. That sounds like a mouthful, but all it means is cells that have been chemically persuaded to roll back the clock, to the time when they were cells that could become any other type of cell.

Could you use the same method to transform a skin cell into an egg? There’s no reason why not, in theory at least. In fact, quite a lot of progress has been made towards producing sperm (which are easier) from stem cells. The reason given for attempting this has been as a possible treatment for infertility, but, once it’s possible to use skin cells from a man to make sperm, it might not be too hard to use cells from a woman to achieve the same thing. Eventually, the skin-to-egg method is likely to be possible as well, although making an egg is a much harder task. Anna and Eve might get first crack at this technology, it seems. Still, the technical challenges are just that — challenges, which can undoubtedly be overcome.

However, it’s likely that there will be objections raised to actually making sperm from a woman’s cells or eggs from a man’s. From a practical perspective, it would be very hard to be sure that this could be done safely. Would an embryo formed from such a manufactured egg or sperm grow into a healthy baby? Who knows? There’s no way of testing it without trying, and no guarantee that if this works in animals it will be safe in humans. On the other hand, the history of this type of technology has been that if it is possible then someone, somewhere will give it a go.

Let’s leave Adam and Steve now, and consider a heterosexual couple who are planning a baby. John and Jane have dreams for their yet-to-be-conceived child. They want him or her to be healthy, of course, but perhaps they hope for more than that. Jane, sporty herself, hopes for a child who might become an Olympic athlete. John, raised in a difficult home environment, wishes for a child who will be compassionate and kind. They know that being smart and attractive gives people an advantage in life, so of course they want those things for little Oscar or Sophie as well.

Can they choose? Should they be able to, if they can?

The diagnosis of genetic disease is unique in that it can be done not only before a person is born — prenatal diagnosis — but before an embryo is even implanted in the mother’s womb. Pre-implantation genetic testing (PGT) isn’t just for mitochondrial conditions; we can test a bundle of cells too small to even see without a microscope, and say, ‘If this embryo grows into a person, one day he or she will have Huntington disease’ (or any of thousands of other conditions).

For PGT, in-vitro fertilisation is used to make as many embryos as possible (which is also the usual goal when IVF is done to treat infertility). The embryos are allowed to grow for a few days until they form a little ball of cells. In the most delicate of medical procedures, a few cells are sucked out, and DNA from those cells is tested for a specific genetic condition. Then, an embryo that is known not to be affected by the condition is implanted. This way, the parents can be sure from the beginning that their child will not be affected.

We can do PGT for single-gene conditions only if the exact genetic cause of a condition in a family is known. Even so, it’s pretty common for the idea of ‘designer babies’ to be raised when PGT is discussed. This is obvious nonsense in relation to the way the testing is actually used in most countries, but it’s easy to understand where the idea comes from. If you can choose to avoid a bad outcome such as a fatal genetic disease, could you select in favour of something you think is desirable in your child, such as intelligence, height, beauty, or sporting ability?

The closest this gets to a ‘yes’ is using the technology to choose to have a boy or a girl. In some countries, boys are favoured over girls, to the extent that those who can afford it sometimes use PGT to choose to have a boy. It seems likely this has been widely done across the world. In some countries, the practice is banned — even those, like Australia, where there has not been any particular bias towards one sex or the other among parents who seek selection. Using PGT for ‘family balancing’ is frowned on, and even banned, for reasons I have never understood. If you have two boys and would like your third child to be a girl, there would be no obvious harm to that girl or to her brothers, and no apparent damage to society at large from that decision. Would she be a ‘designer baby’? Only by the very broadest of definitions. As we shall see in chapter 9, the way that characteristics like height and intelligence are inherited makes it unlikely that we will be able to pick the ‘best’ embryo from those that are available. Even if we could, we wouldn’t have ‘designed’ the baby, we would just have selected one who could have been naturally conceived anyway.

So — if you can’t choose your preferred embryo in any meaningful way, can you manufacture an embryo to suit? The answer is, a very qualified, yes. The genie is out of the bottle, but this genie is not necessarily to be trusted.

Genie: Okay, Dave, you get one wish. Use it wisely!

Dave: I wish I was rich.

Genie: You got it! Enjoy!

Rich: Hey, wait a minute …

It has long been possible to modify the genes of a living creature. We talk about ‘making a mouse’ or ‘making a fly’ (or a fish, or a worm) in order to study what happens when the function of a particular gene is altered. This can include knocking a gene out, so it doesn’t work at all, or adding a specific mutation to a gene, or just putting in extra copies. You can put genes from one creature into another — a famous example is putting the gene that makes some types of jellyfish fluorescent into other animals, so that you get a rabbit that glows green when you shine the right sort of light onto it. There are practical applications to the jellyfish protein, it’s not just a genetic engineering party trick. For instance, if you want to work out if a specific protein is needed in a particular developing tissue, you can splice in the jellyfish protein next to the gene you’re interested in, and then see which bits of an embryo glow green. The green glow marks out very precisely where the protein you’re interested in can be found.

Genetically modified organisms are already economically important — mainly in crops, of which there are many, but there are also a number of animals that have been modified and will probably find their way to your plate soon, if they haven’t yet. Could you modify humans? Of course you could — what works for one mammal works for another. But what type of genetically modified human (GMH) would you make?

One option might be to create a GMH with big muscles. There’s a gene for a protein called myostatin, which acts as an off-switch to muscle growth. Some people aspire to look like Arnold Schwarzenegger in the 1970s, but, from an evolutionary perspective, there’s an advantage to keeping the growth of muscles in check. Making muscles takes nutrients, and then once you have them, you need extra food to keep them going. Limiting muscle growth is all about resource management: efficiency demands that we have enough muscle, but not too much. But if resources are no obstacle … well, there’s a breed of cow called the Belgian Blue that has enormous muscles because it has faulty myostatin. That’s an advantage (although not to the cow) if your plan is to eat that muscle. Would it be helpful otherwise?

There’s at least one human being reported who lacks myostatin. When last described, 14 years ago, he was an extraordinarily heavily muscled and strong four-year-old. Presumably, he is now an extraordinarily heavily muscled and strong young man. So, you might say — what’s the problem? Let’s get cracking and create an army of Olympic weightlifters!

There are two problems, and both relate to safety. When you modify an organism’s genes, there’s a chance that things might go wrong. In the process of knocking out myostatin, you might inadvertently take out something else you would rather have left alone. You could wind up a with a child who is very muscular, but who has a high risk of developing bowel cancer at a young age, or who is born with severe epilepsy. The second problem is that even if everything goes perfectly, we don’t know what the long-term effects might be for people with no myostatin. Perhaps they will remain healthy all their lives. But one healthy four-year-old gives you no grounds for confidence about that. Belgian Blue cattle are somewhat fragile — they struggle in harsh climates, they have difficulty giving birth, and their fertility is lower than that of other breeds. It may be that some of those problems are separate from their enhanced muscles, but there is no way of knowing if a GMH lacking in myostatin might also have long-term health problems.

Or maybe you are after basketballers. Growth hormone, as its name implies, makes you grow. So how about a GMH who has extra growth hormone? Well, we already know how this one turns out. It would indeed produce people who are tall — very tall — but they would very definitely have other health problems as a result. We know this because there is a naturally occurring version of this GMH as well. André Roussimoff, known as André the Giant (who so superbly played Fezzik in The Princess Bride) had this condition: acromegaly. His over-production of growth hormone was caused by a tumour, rather than by any genetic difference, but the effect would be the same either way. Indeed, Roussimoff was tall (224 cm/7 ft 4 in.), and strong, a professional wrestler. But with his size and strength came overgrowth of his facial features and a host of health problems, leading to his death aged just 46.

In late 2018, it was announced that a Chinese scientist called He Jiankui had used the CRISPR technology, a powerful tool for editing genetic material, to change the genetic make-up of two babies in an attempt to make them resistant to infection by HIV. Later, it was revealed that a third baby had also been born. Dr He’s stated goal was to introduce a specific change in the gene CCR5. This gene codes for a protein that sits on the surface of some white blood cells, and is exploited by HIV to get into and infect those cells. The particular change Dr He was trying to introduce is fairly common in Europeans but absent from Asians; people with two copies of the changed protein (about 1 per cent of Europeans) are resistant to HIV infection.

The couples recruited for He’s research were chosen because, in each case, the father had HIV. On the face of it, this makes He’s intentions seem like they might be justified: he wanted to make babies who would not be affected by their father’s HIV; a purely medical motivation. Except that this would have been medical nonsense: there are already sperm-washing techniques that mean that it’s possible for an HIV-positive man to father a child with an extremely low risk of being born infected by the virus. There’s no medical justification for changing such a child’s DNA to further reduce that small risk. Dr He himself, interviewed in November 2018, stated that his goal was to protect the babies against HIV infection later in life. So what was the point of recruiting HIV-positive fathers? Your guess is as good as mine. He doesn’t seem to have offered any coherent explanation.

Later, it emerged that although He did succeed in modifying the CCR5 gene in the babies, he didn’t manage to introduce the specific change found in Europeans, meaning that there’s no way of knowing for sure whether those will be resistant to HIV — particularly because it seems that the changes only affected some of the babies’ cells, i.e. they are mosaic. There are potential downsides to having CCR5 that doesn’t work normally, with vulnerability to some other viruses being part of the price paid for HIV resistance. It’s not at all clear whether these three babies have been helped or harmed, and that’s assuming that the gene that was targeted is the only gene that was changed in the process.

Sound ethically dubious to you? It did to the Chinese authorities, too, and an investigation revealed that He had acted without proper ethical approvals in place. In 2019, he was jailed for three years and heavily fined for his actions.

This is not to say that there are no conceivable medical reasons why you might edit an embryo’s DNA. Almost always, when there is a specific genetic condition in a family, it’s possible to use PGT to select embryos that are not affected by that condition. Sometimes, though, there are circumstances in which this may not be an option. Suppose, for example, that both parents are affected by the same autosomal recessive condition (perhaps they met in the waiting area for a specialised clinic). Both parents have two faulty copies of the gene in question; neither has a normal copy. This means that every embryo they conceive will also be affected, so there would be no unaffected embryos for PGT to choose. At present, if they want to have a child who is not affected by the same condition, they would have to consider adoption, or the use of donor eggs or donor sperm. Gene editing could open the door for them to have healthy children who are biologically their own. Would that be such a bad thing? It seems possible that with careful preparatory work, likely to take years, this may one day become a standard medical procedure.

Having said that, many would think that safety concerns are not the only objections that should be raised to deliberately changing the genetic make-up of a human being, particularly in a way that can be passed on to that person’s own children. Some would argue that this is playing God, against nature, or otherwise an ethical minefield. Whether you accept that or not, there can be no doubt that deliberately modifying an embryo to create an ‘improved’ human raises extra questions about safety, not only from the procedure but from the changes that you are aiming to make, that seem impossible to answer. Only someone completely unscrupulous would attempt such a thing.

Which means, of course, that, by now, someone, somewhere, has done it already. There is surely a genetically modified superbaby out there in the world. For her sake, and for the sake of her future cousins, I hope my concerns about safety are wrong.