Juan Carlos Izpisua Belmonte has long been fascinated by frogs, fish and salamanders—particularly by their ability to regrow lost limbs. He has spent decades, first in Europe and then at the Salk Institute in the U.S., tracing the genetic and developmental steps that allow these animals to regenerate legs and fins, among other things. But his latest research project has focused more on mammals—specifically humans, pigs, cows and sheep. Building on his earlier work understanding regeneration, Izpisua Belmonte is coaxing human tissue to grow inside animal embryos. The result is what's known as a chimeric embryo—named after a mythic Greek animal known as a chimera, which was part lion, part goat and a little bit of a snake.

There are many technical challenges to creating human-animal chimeric embryos  and the first successful experiments—in mice—were performed only two years ago. No scientific papers about chimeric livestock have, as of yet, been published, so we don't have a lot of detail about work with larger animals. But as word spread about some preliminary results, the National Institutes decided rather abruptly last fall not to fund such research—at least for now.

One far-off future application might be the ability to grow human hearts, kidneys and other organs inside pigs and other farm animals to alleviate the donor shortage for transplantation. Researchers still need to understand a lot of basic developmental biology, however, before they can make good on that promise.

What follows is an edited excerpt from a telephone interview that I conducted with Dr. Izpisua Belmonte:

Scientific American: Why is this happening now?

Juan Carlos Izpisua Belmonte (JCIB): Scientists have tried for the last 15 years, more or less, to educate embryonic stem cells and later induced pluripotent stem cells to become the different cell types that the body is composed of. There have been many, many great advances and a lot of progress but the reality is that nothing has transitioned to the clinical practice today. And the major problem is that the cells that we all scientists have been generating in the Petri dish, they are not functional and they are not functional because they are not mature enough so they don’t reach the final stage of education, so to speak. We don't know what to put in the Petri dish to make them fully functional and fully mature cells.

Science takes a long time and it means that little by little we will get to know more based mainly on discoveries from the developmental biology field what to add to the Petri dish for these cells to mature. But this is like a trial-and-error experiment. We add ABC and let's hope if we get something. Therefore this is something that may take a year or may take 10 years. We really don't know.

So along the same lines, there could be other approaches towards trying to regenerate the cells that the human body loses by disease or aging in our lives. One of them looks at what other animals do that normally have this ability to regenerate their bodies. Like the zebra fish, the salamanders, etc. And the major difference between them and the Petri dish that we have in the lab is the niche. So the environmental setup that these cells have to regenerate is very different inside the body than outside the body. We cannot re-create that niche, those environmental cues. And these animals can.

And from there is where the idea came. So normally the embryo knows how to generate different cells and tissues and 99% of the time the baby is perfect. So why don't we put human embryonic stem cells inside an animal embryo and let the animal do the job for us—since we don’t know—and the niche is there? Probably the animal will be able to adopt those cells and guide and educate the cells like if it was their cells. And this is mainly the concept—having a host, an in vivo host, rather than a Petri dish to educate the cells to become whatever we want.

Now, I don't think that this is going to be easy. I have read many, many you know comments or even reports saying that we are going to have human beings grown in an animal very soon. I think we are far away from that because until say a few months, there was not any publication that could show that a human cell could grow inside the blastocyst of another animal.

All these pluripotent stem cell lines that we have been working with, no one was able to integrate in the blastocyst of a mouse embryo. That means we are really far away. It was just in the last few months that our group and another group has been able now to show the first examples that human cells can colonize the early embryo of another species.

Scientific American: So if the stem cells weren't colonizing the new embryo were they just degenerating and disappearing?

JCIB: Yes. So you put them [the cells] in and one thing that is important . . .  How do I explain this? Let's say if you have a freeway with many cars going on this freeway and then you have an entrance and a new car wants to enter into that freeway, it has to enter at a specific speed, at a specific location—to the right and not to the left—and it has to be in sync with the speed and what is happening on that freeway. So the pluripotent cell that we have now is that car that is entering into the freeway and the freeway is the embryo developing with many cars and many cells going to different places.

So we really need to have a cell that has the property of going at the right speed, of knowing how to interpret the signals that are in this freeway etc. and etc. So if not, there will be an accident. And the accident is when we put these cells in, either the cells die or the embryo dies—one of the two things happens. So if you put many cells, that will create a mess in this freeway and the embryo just gets reabsorbed and doesn’t develop. Or if you don't put too many, it's like the other cells don't care and they just put them aside and they die.

So it has to be a really a choreography process to have the cells being integrated. For that you need to have the right cell type. And in the lab we have been getting embryonic stem cells that we really don't know at what stage of their development they are. They could be very early. They could be very late. So we need to find the right one to put at the right time that matches the embryo—the stage at which we are putting the cells.

Scientific American: So you have certain growth factors or something that you add to the cells?

JCIB: Yes. We have now been developing new media, new conditions to grow different cell types. And in fact we have now one paper coming out that it shows that we cannot only generate all the cell types of the embryo but these cells can also generate the placenta, which is the other part of the procedure to have a live organism. You can generate all the cells but if you don't generate the placenta, nothing will happen. And this is by changing the composition of the media.

Now, still even if you put the cells there and they are perfect and they integrate, where do they go? Normally, they will just go randomly to the heart, to the liver and will make a chimera of many different organs. And I don’t know if that’s going to be useful. Because what we want to create is an organ inside an animal that can be then taken out and used for transplantation. So you need to do a trick in that embryo. And the trick now that it is most logical one to do is to try to eliminate from the host embryos the cells that you want to create, the organ or the tissue that you want to create.

So if you want to create a pancreas, you have first to eliminate the pancreas from that embryo so that the human cells normally they will go everywhere but somehow if there is no pancreas, the embryo is trying to compensate for that absence of pancreas and may direct the cells to go to the pancreas location and then make a human pancreas. So how is that possible?

In the last year the gene editing technologies have advanced quite a lot as you know and we now can edit the genome of many animals including the pig which we are now thinking of as the host animal for human cells. So we can eliminate genes from the pig genes that are key to form specific organs. So for instance, the elimination of one particular gene—PDX1—eliminates the formation of the pancreas. So if we do gene editing in these pigs and eliminate the pancreas and at the same time we put human cells in that embryo, it is possible that cells will go everywhere but they will have an affinity to go to a place where it is needed. Otherwise they will not develop. And so that's the other trick that we are using to generate human cells inside another animal.

So where are we now? Right now we have been doing this between mouse and rats to see—they are very close animal species and they take the same time to develop —three weeks until the embryo finally develops and it’s born. And the experiments really have been spectacular. You eliminate say the pancreas gene from the mouse embryo, inject rat cells and you get a mouse with a rat pancreas. Or you get a mouse with an eye from the rat, etc. and etc. We have been looking for different genes for different organs.

So with that idea, we thought. “Great, this is working. Now we need to move to a more complicated experiment.” Human and pig, they are very different. They have been separated for millions of years of evolution. They have a different gestation period. That means that the timing, the synchronization of the car and the freeway is going to be a major challenge. If you put a cell there that is going at a different speed, a different proliferation rate than the host. The host just takes three months to develop a full embryo and the human cells need nine months. So that's also already a problem.

And that's where we are. We have started to do the first experiment and luckily we see that human cells integrate in the pig embryo, not only in the pig embryo, we have seen also in cows and sheep embryos, so that's going fine. And we are now trying to understand where do they go, what is the right moment to put them in?

Another way is what we call cell competition. So there is a phenomenon in biology by which some cells are more fit than others to colonize a specific environment. And that's related to the expression of certain genes. And in fact these genes are also involved in cancer. Cancer cells overexpress these genes and they can colonize many other environments that normal cells will not colonize and they make a tumor.

So based on the information that we have from this field, we are endowing the human cells with properties to be more competent to colonize the embryo because the embryo normally does not like to have cells that are not their cells, so to speak. So we need to try to compensate for that and that's the role of cell competition. This is an area of research that is very interesting now because it could have implications for the cancer field as well and that's where we are more focused now trying to understand how a cell that normally will have some trouble in colonizing another species’ embryo, how can that cell have better abilities to do so?

Scientific American: So this is very, very basic biology?

JCIB: So I feel that there has been a little bit of exaggeration of where we could go with this now. If you look on the Internet you see images of chimeras between human and animal. And I feel that that’s a little bit of exaggeration. It's true that it works very nicely between rat and mouse — just this experimental protocol that I am telling you. It's only a couple of months ago that we have been able to put human cells into another animal. In this case in a mouse and realized that they can differentiate in the three germ layers. The three germ layers are the mesoderm, ectoderm and endoderm that will give rise to the more than 250 different cell types. So that's a major accomplishment I will say. But from there, dreaming that they will generate a functional structure, I think we're going to need time and a lot of luck.

So we need to go for a lot of basic research still. It's my own feeling, of course. There are other people who think that tomorrow we are going to create human organs. And I wish that I am wrong and they are right, but I think it will take time.

Scientific American: How far along have these human-animal chimeras developed?

JCIB: We are entering into an ethical [area]. Because there are some people who think that we shouldn't mix human cells with other animals and there are others who don't care, so to speak. Here in California, we have gone through the different committees and they allow us to have a pig embryo develop for a month. Which is one third of their gestation. At that point you can see already all of the major organ primordia.

There are other countries.  I'm from Spain and Spain has been quite open to this field of stem cell research. And they have allowed us to go until the animal is born. So in theory we could have a pig born with the human organ. It was not easy. Even though Spain is quite open to this stem cell research area, at the same time, Spain is a very Catholic country, so we had to go through the Pope. He very nicely said yes. This is to help people.

Scientific American: The current Pope?

JCIB: Yes. The current Pope. So the Vatican is behind this research and has no problem based on the idea is to help humankind. And in theory all that we will be doing is killing pigs.

One problem and the major problem is that these cells could colonize the brain of the animal in which you put them. And obviously it would not be appropriate to have an animal with neurons from people. Or these cells could colonize the germline so that the sperm or the oocytes of that pig would be human. So to avoid that the government of Spain allowed us to have the pig be born and then immediately after to be sacrificed.

But I was not happy with that. People will think that still you will have an embryo maybe with some neuron contribution. And even though the pig is not born, there are people who believe that that should not be done. So we are devising genetic engineering technology so that if a cell becomes a neuron it is just destroyed in the embryo. Any cell that starts to be taught okay you are going to become a neuron at the moment of the first stages of neurogenesis, we are putting a toxin construct in it so that it will be destroyed by itself. So that will prevent any pig embryos from having human neurons so to speak.

I feel that this will still generate controversy. Many people will think one way and others will think differently, so it is impossible to have a consensus. My feeling is that we still need to better understand these issues of cell competence, of mixing cells in embryogenesis—the rules of development, so to speak. And I am a developmental biologist by background and that is my own interest. It will take a long time to have all these hopes and dreams come true.

Correction: (Jan. 27) The article was updated to make clear that no research articles have so far been published regarding the successful integration of human stem cells into pig, cow or sheep embryos. A successful experiment in mice was conducted two years ago.