Scientists have created synthetic human embryos using stem cells in a groundbreaking advance that sidesteps the need for eggs or sperm. Madeleine Finlay speaks to science correspondent Hannah Devlin about her world exclusive story on this development, what it could mean for medical research, and whether the ethical and regulatory classifications of these embryos are keeping pace with the science. Help support our independent journalism at theguardian.com/sciencepod

Does it matter how our biological lives begin?

These are just some of the huge ethical and philosophical questions raised by a groundbreaking scientific advance announced last week.

Researchers in the US and the UK say they've made a leap forward.

They've created synthetic human embryo-like structures from stem cells that are more advanced than any similar structures made so far.

A group of scientists from the US and UK teaming up to create what is being called the world's first synthetic human embryos made without eggs or sperm.

Researchers say these structures could revolutionize our understanding of the earliest stages of human development, the impact of genetic disorders, and the causes of recurrent miscarriage.

But as the science races ahead, it leaves in its wake fundamental legal and ethical implications.

So today we're asking, how Could synthetic human embryos transform medical research?

From The Guardian, I'm Madeline Finlay, and this is Science Weekly.

Hannah Devlin, you're The Guardian's science correspondent, and last week you broke this huge exclusive, the paper about synthetic human embryos.

It was based on the work of Professor Magdalena Zernica Goertz of the University of Cambridge and the California Institute of Technology.

Now, this hasn't been officially published, but she announced it at a plenary address at the International Society for Stem Cell Research's annual meeting in Boston.

So what this team did was they created essentially lab-grown replicas of embryos of natural human embryos, but they didn't use eggs or sperm and there was no fertilization step involved as you'd normally have.

These were human embryo models or sometimes they're called synthetic embryos that were grown entirely from human stem cells.

So tell me a bit more about exactly what it was that they cultivated in this case.

What does this embryo-like structure resemble?

So they started with embryonic stem cells, human embryonic stem cells, which are the cells that you have at the very earliest stage of development and they've got this amazing capability of turning into any cell type in the body.

And they've already been used a lot in research to kind of turn them into, say, heart cells or neurons to grow bits of tissue in the dish.

But a very new area is trying to get these cells to actually replicate the steps that happen in an embryo.

So not just turning into 1 cell type, but turning into the very earliest structure that you have that would then go on to develop into a baby eventually.

Now I should emphasize that that is not at all their intention.

This is very much done for scientific purposes in this case.

So in this case, they took these embryonic stem cells, they cultivated them in this special culture system.

So it's basically like a rotating bottle with various nutrients and things to encourage the cells to grow that very approximately mimics the environment inside the human womb.

They started by separating the cells into 3 groups and 1 of the groups was going to go on to be the embryo itself and then the other 2 groups were going to become the cells that would form the placenta and the yolk sac.

And so they cultivate these 3 populations of cells for a few days separately and then they put them back together and put them into this culture system and then they allowed it to develop to something equivalent to the 14th day of development of an embryo.

So what they had at the end of it didn't precisely mirror what you have in a natural embryo, But what they did have were the precursors of the cells that would go on to become the placenta and the yolk sac, and then also the precursors of what become the egg or sperm cells in a real embryo, which is 1 of the first things to start to develop in an actual embryo.

So these models have the components that you would see in a human embryo, but could they have developed into full embryos?

I mean, I think in this specific case, these specific embryo models that we broke the story about last week, the answer is probably no, because since then some of the more detailed data has been published and it's clear that they're not a precise mirror of what an actual embryo looks like at 14 days.

But I think there is still a very big question about what the ultimate potential is for these structures.

And just to give an example, last year there was a breakthrough on the same topic using mice cells, which took it really quite a lot further.

And they were able to build these synthetic mice embryos.

It was also Magdalena Zernicka-Gertz's team and another lab based in Israel.

They had the beginning of a brain, beating heart, a gut-like structure, but they did find they couldn't successfully implant these into mice and create a viable pregnancy.

So I think there is a totally open question among scientists about whether these structures might eventually have the theoretical potential to fully develop and whether it's just a kind of technical hurdle that needs to be overcome or whether there's some more fundamental biological hurdle that, you know, something about the process of fertilization, something about those steps, programs the embryo right at the start to know exactly how to develop.

And even scientists don't know the answer to that.

But right now, the plan is certainly not to allow these to develop into full embryos, even if they could.

So why do scientists want to do this work?

1 of the real interests in this is that there is currently a complete black box period in human development, effectively between 14 days where there's a legal cutoff in most countries for cultivating natural embryos in the lab and then around 35 days where you can pick up the course of development on say pregnancy scans or embryos that have been donated.

So there's this window where really not a lot is known about exactly what's happening and it's a really crucial window.

That's a point at which a lot of pregnancies fail, for people who've got fertility issues or recurrent miscarriage, it's really crucial to get better insights into what's happening in that time period and could also provide new insights into the effects of drugs on embryos, on gestational diabetes and how that affects the embryos development.

And then there's sort of more futuristic applications.

So some People have talked about using these systems for regenerative medicine.

The idea that you could actually grow parts of organs even for a patient.

Say for example you had someone who was waiting for a liver transplant, potentially, 1 scientist told me, you could grow a bit of a liver using 1 of these systems and then using a bit of this liver as a graft for the patient as a sort of bridge until they actually got their transplant.

I mean, you know, this sounds very science fiction, but who knows?

These systems themselves weren't really on the horizon a few years ago.

So perhaps some of these more far-reaching applications might come to be a reality.

These models, I mean, how much can we really learn from them at the moment?

If we've got this black box, Do we know how close they are to the real thing?

I think 1 interesting thing is that it will be necessary, if you want to really be confident that they're replicating that process and kind of giving you this window into the black box, you do need to be able to compare them to what's happening in a natural embryo.

So there's an international consensus now that a kind of blanket 14-day rule for natural embryos is not appropriate.

They think it should be judged on a case-by-case basis and there should be some flexibility about allowing scientists to culture natural embryos to a slightly later stage.

And it's interesting you saying that these models now are beginning to get closer potentially to the real thing Because at the same meeting that we heard about the first story in Boston, there was another big story that came out of that.

A team from the University of Cambridge's Gurdon Institute created a model human embryo with a heartbeat and traces of blood.

This was another team based at Cambridge, based in the lab of Professor Azim Sarani.

A member of his team presented work which was also a human embryo model, but which was actually cultivated to have some of the features of a far later stage of development.

It's not a human heart as we would have ourselves with ventricles and complex structure, but it was a cluster of cells that were pulsating and there were red blood cells present as well.

I think a really key difference with this piece of work was unlike the teams that have been working up to this kind of 14-day stage, this latest 1 with the precursor of heart cells didn't contain all the structures that you would have in a natural embryo developing.

So they intentionally didn't include the precursors to placenta cells and the yolk sac.

So I think they felt there was no theoretical chance even that what they were cultivating had the potential to go on and develop far further into a foetus, for example.

So they felt that that put them on really safe ethical ground to maybe cultivate it a bit further and look at some of these structures that develop a little bit further along.

From what you've described it sounds like this field is progressing really quickly and there's some serious competition between the groups.

It's worth saying that Magdalena Zernicka-Gerz has been working on this for decades.

She's really spent her career developing these systems and asking these questions.

And so have many of the scientists involved.

But I think it's absolutely the case that 5 years ago, a lot of the ethical questions that these advances are throwing up were totally hypothetical and suddenly, the science has suddenly outpaced the regulations, guidelines and law in this area.

There is intense competition between the different labs involved in this.

They're all racing to get their research out to be the first over the line.

I think they are responsible and cautious about the ethics around this.

There's a lot of international discussion around what appropriate guidelines should be placed around this research.

But yeah, I mean, there is definitely fierce rivalry and it's a really hot topic.

And in part because of the speed of the research, it's meant that we're catching up in terms of the legal and ethical ramifications of the work.

But where does the current legislation stand on what scientists are allowed to do and what's the advice for scientists doing this work?

The first thing to say is this is not totally unregulated.

So there are extremely clear laws around the fact that you couldn't ever implant these human embryo models.

It would be completely illegal to bring them into the clinic in any way.

And there are also laws and rules around the use of human cells for research.

So I think where the key gap is, is that there are a special set of rules that apply to human embryo research.

Whereas those kind of rules do not apply to these human embryo models.

And now that they're becoming very embryo-like in some cases, There's this big question about exactly which bits of those rules should apply to the embryo model.

Some people think that if they look the same, if they're behaving the same, they should be treated the same in a legal sense.

But there's a spectrum of opinion out there on the best way to regulate this area and it's a kind of life question really.

And so how do you take these conversations forward?

Who needs to be involved in them and to begin drafting legislation or regulations?

Speaking for the UK, there are already a group of scientists who've been working on this, I think informally since last year, but formally since March, to come up with a set of guidelines that they hope the whole community in the UK and possibly beyond will form a consensus around.

So that in the absence of an actual law around this, there'll at least be a guideline that all scientists can agree to adhere to.

They don't want this to be a kind of grey area because it puts labs in a really difficult position if they're each having to navigate what their own sort of risk appetite is for doing this work.

There's also in the UK a potentially big overhaul of our entire fertility legislation underway.

So the fertility watchdog, the HFPA, has drawn up proposed changes to our laws which cover lots of different things around how fertility clinics are regulated, all sorts of other stuff, but this bit of science is 1 of the other things they're looking at and it's quite likely that this will be included in any updates to our legislation and it'll be really interesting to see how they are going to do that because it's an incredibly technical area, it's something where you want the laws to be clear but also to have some flexibility to allow them to take in future advances that might be on the horizon.

So it's going to require really sort of expert piece of work to make sure that they get that right in the law.

Hannah, these models do prompt bigger, more philosophical questions in terms of what makes humans humans and I wonder what your thoughts on that are?

Yeah, they really do and I think it's a really good example of where science can really cause you to question why something's making you feel uncomfortable, why we have an ethical rule around a certain thing and then suddenly you have to kind of figure out precisely what it is about cultivating embryos beyond a certain point that we don't like.

It's absolutely fascinating trying to figure out what these things are.

They're embryo-like, but they're not embryos.

They replicate a lot of the features that we once had when we first developed but they don't have personhood.

Someone on a briefing last week was talking about the concept of insolements and when insolement happens.

And it's not that often that science raises these kind of questions but I think it's really fascinating when it does.

It's going to be so interesting to see what happens next.

Yeah, I'm looking forward to discovering this field over the next few years.

You can find all of her brilliant reporting on theguardian.com.

The producer was me, Madeline Finlay, the sound design was by Tony Onuchukwu, and the executive producer was Ellie Burey.


Speaker 0: This is The Guardian. What makes us human? Does it matter how our biological lives begin? These are just some of the huge ethical and philosophical questions raised by a groundbreaking scientific advance announced last week.

Speaker 1: Researchers in the US and the UK say they've made a leap forward. They've created synthetic human embryo-like structures from stem cells that are more advanced than any similar structures made so far.

Speaker 2: A group of scientists from the US and UK teaming up to create what is being called the world's first synthetic human embryos made without eggs or sperm.

Speaker 0: Researchers say these structures could revolutionize our understanding of the earliest stages of human development, the impact of genetic disorders, and the causes of recurrent miscarriage. But as the science races ahead, it leaves in its wake fundamental legal and ethical implications. So today we're asking, how Could synthetic human embryos transform medical research? And can the law keep up? From The Guardian, I'm Madeline Finlay, and this is Science Weekly.

Hannah Devlin, you're The Guardian's science correspondent, and last week you broke this huge exclusive, the paper about synthetic human embryos. It was based on the work of Professor Magdalena Zernica Goertz of the University of Cambridge and the California Institute of Technology. Now, this hasn't been officially published, but she announced it at a plenary address at the International Society for Stem Cell Research's annual meeting in Boston. Tell me more.

Speaker 3: So what this team did was they created essentially lab-grown replicas of embryos of natural human embryos, but they didn't use eggs or sperm and there was no fertilization step involved as you'd normally have. These were human embryo models or sometimes they're called synthetic embryos that were grown entirely from human stem cells.

Speaker 0: So tell me a bit more about exactly what it was that they cultivated in this case. What does this embryo-like structure resemble?

Speaker 3: So they started with embryonic stem cells, human embryonic stem cells, which are the cells that you have at the very earliest stage of development and they've got this amazing capability of turning into any cell type in the body. And they've already been used a lot in research to kind of turn them into, say, heart cells or neurons to grow bits of tissue in the dish. But a very new area is trying to get these cells to actually replicate the steps that happen in an embryo. So not just turning into 1 cell type, but turning into the very earliest structure that you have that would then go on to develop into a baby eventually. Now I should emphasize that that is not at all their intention.

This is very much done for scientific purposes in this case. So in this case, they took these embryonic stem cells, they cultivated them in this special culture system. So it's basically like a rotating bottle with various nutrients and things to encourage the cells to grow that very approximately mimics the environment inside the human womb. They started by separating the cells into 3 groups and 1 of the groups was going to go on to be the embryo itself and then the other 2 groups were going to become the cells that would form the placenta and the yolk sac. And so they cultivate these 3 populations of cells for a few days separately and then they put them back together and put them into this culture system and then they allowed it to develop to something equivalent to the 14th day of development of an embryo.

So what they had at the end of it didn't precisely mirror what you have in a natural embryo, But what they did have were the precursors of the cells that would go on to become the placenta and the yolk sac, and then also the precursors of what become the egg or sperm cells in a real embryo, which is 1 of the first things to start to develop in an actual embryo.

Speaker 0: So these models have the components that you would see in a human embryo, but could they have developed into full embryos?

Speaker 3: I mean, I think in this specific case, these specific embryo models that we broke the story about last week, the answer is probably no, because since then some of the more detailed data has been published and it's clear that they're not a precise mirror of what an actual embryo looks like at 14 days. But I think there is still a very big question about what the ultimate potential is for these structures. And just to give an example, last year there was a breakthrough on the same topic using mice cells, which took it really quite a lot further. And they were able to build these synthetic mice embryos. It was also Magdalena Zernicka-Gertz's team and another lab based in Israel.

They had the beginning of a brain, beating heart, a gut-like structure, but they did find they couldn't successfully implant these into mice and create a viable pregnancy. So I think there is a totally open question among scientists about whether these structures might eventually have the theoretical potential to fully develop and whether it's just a kind of technical hurdle that needs to be overcome or whether there's some more fundamental biological hurdle that, you know, something about the process of fertilization, something about those steps, programs the embryo right at the start to know exactly how to develop. And even scientists don't know the answer to that.

Speaker 0: But right now, the plan is certainly not to allow these to develop into full embryos, even if they could. So why do scientists want to do this work? What's the aim of these models?

Speaker 3: 1 of the real interests in this is that there is currently a complete black box period in human development, effectively between 14 days where there's a legal cutoff in most countries for cultivating natural embryos in the lab and then around 35 days where you can pick up the course of development on say pregnancy scans or embryos that have been donated. So there's this window where really not a lot is known about exactly what's happening and it's a really crucial window. That's a point at which a lot of pregnancies fail, for people who've got fertility issues or recurrent miscarriage, it's really crucial to get better insights into what's happening in that time period and could also provide new insights into the effects of drugs on embryos, on gestational diabetes and how that affects the embryos development. And then there's sort of more futuristic applications. So some People have talked about using these systems for regenerative medicine.

The idea that you could actually grow parts of organs even for a patient. Say for example you had someone who was waiting for a liver transplant, potentially, 1 scientist told me, you could grow a bit of a liver using 1 of these systems and then using a bit of this liver as a graft for the patient as a sort of bridge until they actually got their transplant. I mean, you know, this sounds very science fiction, but who knows? These systems themselves weren't really on the horizon a few years ago. So perhaps some of these more far-reaching applications might come to be a reality.

Speaker 0: These models, I mean, how much can we really learn from them at the moment? If we've got this black box, Do we know how close they are to the real thing?

Speaker 3: That's a really good question. I think 1 interesting thing is that it will be necessary, if you want to really be confident that they're replicating that process and kind of giving you this window into the black box, you do need to be able to compare them to what's happening in a natural embryo. So there's an international consensus now that a kind of blanket 14-day rule for natural embryos is not appropriate. They think it should be judged on a case-by-case basis and there should be some flexibility about allowing scientists to culture natural embryos to a slightly later stage.

Speaker 0: And it's interesting you saying that these models now are beginning to get closer potentially to the real thing Because at the same meeting that we heard about the first story in Boston, there was another big story that came out of that. A team from the University of Cambridge's Gurdon Institute created a model human embryo with a heartbeat and traces of blood. So what did they manage to do?

Speaker 3: This was another team based at Cambridge, based in the lab of Professor Azim Sarani. A member of his team presented work which was also a human embryo model, but which was actually cultivated to have some of the features of a far later stage of development. So he described it as a heartbeat. It's not a human heart as we would have ourselves with ventricles and complex structure, but it was a cluster of cells that were pulsating and there were red blood cells present as well. I think a really key difference with this piece of work was unlike the teams that have been working up to this kind of 14-day stage, this latest 1 with the precursor of heart cells didn't contain all the structures that you would have in a natural embryo developing.

So they intentionally didn't include the precursors to placenta cells and the yolk sac. So I think they felt there was no theoretical chance even that what they were cultivating had the potential to go on and develop far further into a foetus, for example. So they felt that that put them on really safe ethical ground to maybe cultivate it a bit further and look at some of these structures that develop a little bit further along.

Speaker 0: From what you've described it sounds like this field is progressing really quickly and there's some serious competition between the groups. Is that what you perceive? Is it going quite fast

Speaker 3: as an outsider? It feels like that. It's worth saying that Magdalena Zernicka-Gerz has been working on this for decades. She's really spent her career developing these systems and asking these questions. And so have many of the scientists involved.

But I think it's absolutely the case that 5 years ago, a lot of the ethical questions that these advances are throwing up were totally hypothetical and suddenly, the science has suddenly outpaced the regulations, guidelines and law in this area. There is intense competition between the different labs involved in this. They're all racing to get their research out to be the first over the line. I think they are responsible and cautious about the ethics around this. They take it very seriously.

There's a lot of international discussion around what appropriate guidelines should be placed around this research. But yeah, I mean, there is definitely fierce rivalry and it's a really hot topic.

Speaker 0: And in part because of the speed of the research, it's meant that we're catching up in terms of the legal and ethical ramifications of the work. But where does the current legislation stand on what scientists are allowed to do and what's the advice for scientists doing this work?

Speaker 3: The first thing to say is this is not totally unregulated. So there are extremely clear laws around the fact that you couldn't ever implant these human embryo models. It would be completely illegal to bring them into the clinic in any way. And there are also laws and rules around the use of human cells for research. So I think where the key gap is, is that there are a special set of rules that apply to human embryo research.

It's all extremely tightly monitored. Whereas those kind of rules do not apply to these human embryo models. And now that they're becoming very embryo-like in some cases, There's this big question about exactly which bits of those rules should apply to the embryo model. Some people think that if they look the same, if they're behaving the same, they should be treated the same in a legal sense. But there's a spectrum of opinion out there on the best way to regulate this area and it's a kind of life question really.

Speaker 0: And so how do you take these conversations forward? Who needs to be involved in them and to begin drafting legislation or regulations?

Speaker 3: Speaking for the UK, there are already a group of scientists who've been working on this, I think informally since last year, but formally since March, to come up with a set of guidelines that they hope the whole community in the UK and possibly beyond will form a consensus around. So that in the absence of an actual law around this, there'll at least be a guideline that all scientists can agree to adhere to. And they want that. They don't want this to be a kind of grey area because it puts labs in a really difficult position if they're each having to navigate what their own sort of risk appetite is for doing this work. There's also in the UK a potentially big overhaul of our entire fertility legislation underway.

So the fertility watchdog, the HFPA, has drawn up proposed changes to our laws which cover lots of different things around how fertility clinics are regulated, all sorts of other stuff, but this bit of science is 1 of the other things they're looking at and it's quite likely that this will be included in any updates to our legislation and it'll be really interesting to see how they are going to do that because it's an incredibly technical area, it's something where you want the laws to be clear but also to have some flexibility to allow them to take in future advances that might be on the horizon. So it's going to require really sort of expert piece of work to make sure that they get that right in the law.

Speaker 0: Hannah, these models do prompt bigger, more philosophical questions in terms of what makes humans humans and I wonder what your thoughts on that are?

Speaker 3: Yeah, they really do and I think it's a really good example of where science can really cause you to question why something's making you feel uncomfortable, why we have an ethical rule around a certain thing and then suddenly you have to kind of figure out precisely what it is about cultivating embryos beyond a certain point that we don't like. It's absolutely fascinating trying to figure out what these things are. They're embryo-like, but they're not embryos. They grow from human cells. They replicate a lot of the features that we once had when we first developed but they don't have personhood.

Someone on a briefing last week was talking about the concept of insolements and when insolement happens. And it's not that often that science raises these kind of questions but I think it's really fascinating when it does.

Speaker 0: Hannah, thank you so much. It's going to be so interesting to see what happens next.

Speaker 3: Yeah, I'm looking forward to discovering this field over the next few years.

Speaker 0: Thanks again to Hannah Devlin. You can find all of her brilliant reporting on theguardian.com. And that's it for today. The producer was me, Madeline Finlay, the sound design was by Tony Onuchukwu, and the executive producer was Ellie Burey. We'll be back on Tuesday.

See you then. This is The Guardian.

See all Science Weekly transcripts on Applepodcasts

Synthetic human embryos: can the law keep pace with the science?