Konrad Hochedlinger studied genetics at the University of Vienna. He is now a postdoctoral researcher at the Whitehead Institute of Biomedical Research at the Massachusetts Institute of Technology in the USA. His research interests include embryonic and adult stem cells, nuclear transfer, epigenetic reprogramming and cancer biology. He has shown that it is possible to clone a mouse from mature cells of the immune system, and that embryonic stem cells can cure a mouse of an immune system disease. He has also cloned mice from cancer cells, thus showing that cancer cells can be reprogrammed.

Nuclear transfer to create stem cells
by Dr Konrad Hochedlinger

In the next 30 or 40 minutes or so I would like to give you an update on what we know about nuclear transfer, its use to derive embryonic stem cells, and the potentials these two technologies may have for cell therapy and medicine.

(Click on image for a larger version)

I put up this slide first just to remind myself, and you, that mammalian cloning is actually a very recent discovery. It has been less than 10 years since the cloning of Dolly, the first cloned mammal. The cloning of Dolly, and of many other mammals thereafter, raised a set of very complex questions, some of which I would like to discuss today in my lecture.

For example, it raised major scientific questions about how and why cloning works at all. Then it also raised practical questions. Can we use nuclear transfer in basic research? And can we use it in medicine, for cell therapy? Lastly, ethical questions were raised. For example, should we clone humans at some point?

In my talk I will address only the first two questions: what are the scientific and practical applications of nuclear transfer technology?

(Click on image for a larger version)

I will start by giving you a few explanations of what cloning is about, what the differences are between reproductive cloning and therapeutic cloning, and put this into context with normal development.

This slide shows you how normal development begins. You would all know that at fertilisation a haploid sperm genome fuses or combines with a haploid oocyte genome to form a diploid embryo called a zygote embryo. This then undergoes cleavage divisions, as we have heard this morning from Martin Pera, to form the blastocyst-stage embryo. This blastocyst then implants in the uterus of its mother, to grow into a fetus and eventually into a newborn mouse or individual.

(Click on image for a larger version)

During ‘reproductive’ cloning, you take the diploid nucleus from an adult cell of a mouse or an individual, and place it into an egg whose own DNA has already been removed. Then you create a reconstructed oocyte which basically resembles the zygote-stage embryo. It can also undergo cleavage divisions to form a cloned blastocyst.

The intent of reproductive cloning is to produce a newborn mouse or individual. So what you need to do is to transfer the cloned blastocyst into the uterus of a recipient female, where it can grow into a fetus and eventually into a newborn clone.

(Click on image for a larger version)

The other type of cloning is ‘therapeutic’ cloning, or 'somatic cell nuclear transfer' (SCNT). Here the intention is not to create a newborn clone but instead to derive embryonic stem cells which can be used for cell therapy.

You basically start out the same way, but instead of taking a normal cell you take the nucleus from a patient’s cell and inject it into an oocyte, to create a cloned embryo which then grows into a blastocyst-stage embryo. This embryo is never implanted in the uterus, as it would be during reproductive cloning. Instead, you put it in a petri dish, which allows you to extract embryonic stem cells – and, as we have heard in the previous talks, embryonic stem cells have the potency to give rise to potentially any cell type of the body so they can be differentiated into whatever cell type is affected in the patient, to be used for cell therapy.

[Quicktime image unavailable]

Technically, nuclear transfer involves three steps, the first of which is enucleation. Here you are looking at a mouse oocyte. You see the outer eggshell, called the zona pellucida, and in the centre is the cytoplast containing the chromosomes, which are aligned on a spindle. For enucleation we simply drill through this eggshell with a needle, and suck in a little bit of cytoplasm containing the chromosomes which are aligned on the spindle. It is sometimes hard to see the nucleus, but the opaque area you see here is basically the metaphase plate, where the chromosomes are located at this point of development.

This egg is now enucleated.

[Quicktime image unavailable]

The second step is the preparation of the nucleus for transfer. Here we choose a needle which is slightly smaller than the cell itself that we are going to pick up. It allows you to rupture the membrane and the cytoplasm, and to get rid of these two components so that you end up with the bare nucleus in your needle, ready for nuclear transfer. (You don’t want all these other components in the oocyte.)

The film sequence in this slide shows that we pick up the cell, get rid of the cytoplasm and membranes – which are then discarded – and end up with only the nucleus, which is then ready for nuclear transfer.

[Quicktime image unavailable]

This is the last step, nuclear transfer itself. You see here the previously enucleated eggs. And the needle, which is now loaded with the nuclei which we have just prepared, comes across from the right.

Again we drill through the zona, introduce a hole into the membrane of the egg – so now we are physically inside the oocyte – deposit the nucleus, and withdraw the needle. This egg is now reconstructed. It is now a cloned embryo.

I will now show you, using a movie taken in real time, the speed at which we do nuclear transfer in mice. So, again, we invaginate the membrane, introduce a hole, and deposit the nucleus. This embryo is now a reconstructed oocyte.

(Click on image for a larger version)

You need now to activate those reconstructed embryos artificially, after which they will start dividing into, first, a two-cell embryo, then into a four-cell embryo, an eight-cell embryo, the morula, and eventually, after about four days in mouse, into a blastocyst. If you transplant them or transfer them into the womb of a female during reproductive cloning, they will grow into a newborn clone. If you place them in a petri dish they will grow into an embryonic stem cell line.

(Click on image for a larger version)

To date, a lot of different mammalian species have been cloned. Some are shown here: in addition to Dolly the sheep, cows, cats, pigs, mice and a couple of other species have been successfully cloned, using this technology.

(Click on image for a larger version)

What do we know about reproductive cloning?

(Click on image for a larger version)

The first and most obvious observation was that it is a very inefficient and error-prone process. If you plot the survival rate of cloned embryos against the days of pregnancy, which in mouse is basically three weeks, you can see that about half of the cloned embryos are lost even before the embryo implants in the uterus, and another 47–49 per cent die after implantation. So you basically end up with only 1–3 per cent of the cloned embryos developing to full term. That is highly inefficient.

(Click on image for a larger version)

Many of the clones that do survive are not normal. Many succumb to a phenotype called ‘large offspring syndrome’. You see here a newborn mouse which, compared with a control mouse, is largely overgrown. And the placenta is about four to five times as big as a control placenta.

(Click on image for a larger version)

Even the clones that do survive to adulthood can show abnormalities later in life. To give you an example: mice that had been cloned from Sertoli cells – some of the testes support cells – were shown to die prematurely, compared with controls. (We have here a survival curve.) They also show liver abnormalities and often develop tumours.

Likewise, mice cloned from cumulus cells, for example – the egg lining cells and nourishing cells – become obese later in life. An example is shown here of such an obese mouse, compared with two normal litter mates.

(Click on image for a larger version)

These abnormalities that we see by physical inspection are also reflected at the molecular level. Using gene arrays, where you basically look at the activation and suppression of genes within the entire genome, we can look at what genes are aberrantly activated in clones, as indicated here by the green and red stripes. It turns out that about 4 per cent of all mouse genes are actually abnormally activated in cloned embryos. In the mouse, 4 per cent means a couple of hundred genes that are not expressed normally in these mice.

In addition, up to 50 per cent of so-called imprinted genes are abnormally activated in cloned embryos. Imprinted genes are a special subclass of genes which are particularly vulnerable to environmental stress – for example, the culturing of cloned embryos from the zygote stage to the reconstructed embryo stage or to the blastocyst stage. The misregulation of these genes might, in fact, be the reason why we observe the large offspring syndrome in many cloned species.

You might ask yourself, as we asked ourselves, why cloning is so inefficient, and why it results in all these abnormalities. To discuss that, I first need to explain epigenetic gene regulation to you.

What is epigenetics? Epigenetic modifications can be defined as reversible chemical modifications of DNA, or proteins that are associated with DNA, which influence the readability of genes in different tissues or the readability of genes during development.

(Click on image for a larger version)

The best way to explain epigenetics is to give you an analogy. Let us look at the first of the five ‘sentences’ on this slide: ‘To be or not to be, – that is the question – . This sentence makes sense because it is fully formatted. It has punctuation, it has spaces, and it has lower case and capital letters, making it readable. So, it is fully formatted.

If you were to unformat this sentence by removing the modifications – punctuation, spaces and things like that – you would end up with text such as appears here in the second paragraph. It does not make sense at all to the reader. This is exactly the problem that we are facing during nuclear transfer.

What is required during nuclear transfer is the resetting of such a fully formatted state back into an unformatted ground state – that is, the state of an adult cell needs to be reformatted back into the state of an embryonic cell. We know that this is an inefficient process, and often we see only partial reformatting or, as we call it, ‘reprogramming’ from a fully formatted to a less formatted or an unformatted setting.

So, again in the case of our text, this would mean that in one clone you would reformat only the last three of the five ‘sentences’; the first two would remain formatted. And in another clone, it would be exactly the opposite.

(Click on image for a larger version)

I will give you just one example of an epigenetic modification: DNA methylation. (This is the only formula I am going to bug you with.) DNA methylation is a very prominent example of epigenetic modifications of DNA, which is basically just the addition of a methyl group at the hydrogen position of the nuclear base cytosine, one of the four elements of DNA. It results in the modification of cytosine into 5-methyl-cytosine.

(Click on image for a larger version)

In other words, the problem of nuclear cloning is a problem of faulty epigenetic reprogramming. What does this mean specifically for a donor cell?

If you think of a donor cell – let’s say the mammary gland cell that gave rise to Dolly the sheep – in this donor cell certain genes were active: the genes important for milk production, for example. Other genes that are important for embryonic development, however, are silent. You don’t need them for milk production.

What you now need to do, upon transplantation of this nucleus into the oocyte during cloning, is to reactivate, very rapidly, embryonic genes to allow embryonic and fetal development. At the same time you need to silence all the tissue-specific genes. You don’t need milk production in an embryo – for obvious reasons. We know that methylation plays an important role in this process, and many clones show aberrant erasure and re-establishment of these methylation marks, which are responsible for the activation of these genes.

That explains why very few clones actually survive to birth, and why many of the surviving clones are abnormal.

To summarise: faulty reprogramming is, in a way, a biological barrier. And this biological barrier might, in fact, preclude the generation of ‘normal’ cloned individuals. And all the cloned mammals that have been produced so far may, in fact, not be completely normal.

One important aspect is that all of these abnormalities that I have described to you are due solely to epigenetic abnormalities. They are not due to genetic abnormalities, which would be mutations, basically, translocations, for example, any irreversible change of DNA. This is because the offspring of cloned animals are always normal. So once you pass your genes and chromosomes through the germline again, the abnormal epigenetic modifications are properly erased and also properly re-established, in that the sperm and oocyte genomes are fully functional again and the offspring are fine.

This is what is known about reproductive cloning.

(Click on image for a larger version)

I would now like to switch to the potential use of nuclear transfer technology for therapy and medicine.

Before going into that, I would just like to briefly review what is known about transplantation medicine in general, and what the limitations are.

We know that the transplantation of normal cells into patients represents a valid strategy for treating diseases such as diabetes, Parkinson’s, blood disorders, liver disorders and many more. The donor cells so far have come from either aborted fetuses or cadavers. This poses a huge problem, as you can imagine, firstly because the donor cells are recognised as foreign by the patient’s immune system and therefore rejected by the patient’s immune system, so you need to permanently administer immunosuppressive drugs, which have a lot of side effects. Secondly, these donor cells are not readily available and the quality control of these cells is also very poor.

So therapeutic cloning or somatic cell nuclear transfer might be a potential solution to at least some of these problems of transplantation medicine. This is because nuclear transfer derived embryonic stem cells would, first of all, provide an inexhaustible source of replacement tissue for therapy. Second, these cells would be tailored to the needs of the patient, so  immunosuppressive treatment would not be needed for transplanted cells.

This all sounds very nice in theory, but does it also work in practice?

(Click on image for a larger version)

To address this question, our lab has shown, in a sort of proof of principle experiment, that you can indeed combine nuclear transfer with gene and cell therapy to treat a genetic disorder in a mouse model. I will briefly walk you through the individual steps of this experiment, which are also the basics of therapeutic cloning.

We started by picking a mouse model that carried a mutation in the Rag2 gene. These mice cannot produce any mature immune cells: any B- or T-cells. This is a disorder that is very similar to the human ‘bubble baby’ syndrome. What we did as the first step of therapeutic cloning was to culture somatic cells from the tail. So we cultured skin cells as a first step, and then performed nuclear transfer of these cells into enucleated oocytes as a second step.

(Click on image for a larger version)

The third step then was to produce cloned blastocysts and derive autologous embryonic stem cells from these blastocysts. Since these embryonic stem cells, of course, still carry the Rag2 mutation of the donor mouse in their genome, we then had to perform gene therapy. We had to fix one of the mutant alleles in the embryonic stem cells by gene targeting, and once this was done, the last and really most challenging step was to drive differentiation of these embryonic stem cells into the cells that were affected in the patient: immune precursor cells, or bone marrow stem cells, if you will.

(Click on image for a larger version)

Once this hurdle was overcome, we transplanted those differentiated bone marrow cells back into our sick mouse, where we saw at least partial restoration of the phenotype.

(Click on image for a larger version)

You could extrapolate this to many other tissues and diseases, because embryonic stem cells can, potentially, give rise to any cell type of the body. So, for example, you could differentiate those cells into dopaminergic neurons to treat Parkinson’s, or into skin cells to treat burn victims, into pancreatic islet cells to treat diabetes, into hepatocytes to treat hepatitis, and so on.

But all the data I have shown you so far were obtained in the mouse model. Before thinking about moving into a human setting, there are a couple of caveats or limitations that need to be overcome first.

One of them is, as Martin Pera pointed out this morning, the need for a large number of human recipient oocytes for nuclear transfer. Another question, which was not resolved until very recently, was whether somatic cell nuclear transfer would actually be feasible at all in humans. A few years back there were some indications that it might not work.

(Click on image for a larger version)

So, to answer the second question first, a group in Korea showed last year that somatic cell nuclear transfer in humans is at least technically feasible. They were the first ones to derive a pluripotent human embryonic stem cell line derived from a cloned blastocyst.

So, in theory, somatic cell nuclear transfer can work in humans,

With respect to the other limitation, the human oocytes, there might also be a way to get around it. A group in the United States showed one or two years ago that mouse embryonic stem cells have actually the potential to give rise to oocyte-like cells in a culture dish if they are exposed to certain growth factors. So it may, in fact, be possible to produce an unlimited number of recipient eggs, simply in tissue culture, when human embryonic stem cells are used. But this still needs to be shown. It has not yet been repeated with human ES cells.

All these data I have shown you so far raise yet another set of complex questions, which I would like to address now.

First, do we think that reproductive cloning could be made safe at some point in the future? Another issue is: does the faulty reprogramming that we observe in cloned embryos pose any problem for the therapeutic application of somatic cell nuclear transfer? And, lastly: what is actually the difference between embryonic stem cells derived from an in vitro fertilised embryo and embryonic stem cells derived from a cloned embryo? Are there any differences?

To the first question: can reproductive cloning be made safe in the future? In our opinion the answer is no. In addition to the technical problems, there are serious biological barriers or limitations, because the two parental genomes are differently epigenetically modified during germ cell formation – so, sperm cell and egg maturation. And they remain epigenetically distinct in the adult. You simply cannot recapitulate that by placing an adult donor cell into mouse oocytes during nuclear transfer, but only once you pass the germline again. So only if you go into the next generation, you get complete reprogramming, which at this point is technically not feasible to be reproduced in vitro.

Now to the second question: does faulty reprogramming pose a problem for the use of somatic cell nuclear transfer in therapy? And here again the answer is no.

First of all, during therapeutic cloning there is never a fetus formed – the stage at which most of the abnormalities are manifested, for example the large offspring syndrome.

Secondly, the process of embryonic stem cell derivation is a selection process, where only functional, fully reprogrammed cells will grow into an embryonic stem cell line and nonfunctional cells will be selected against.

We come now to the last point: the difference between embryonic stem cells derived from an in vitro fertilised embryo and a cloned embryo.

IVF was used to produce most, if not all, of the human embryonic stem cell lines that have been used so far in laboratories all over the world. The intent of in vitro fertilisation is to generate a new person. It involves the creation of new life – by a unique genetic combination. And these embryos produced by in vitro fertilisation have a very high potential to generate a normal baby when transferred into a female.

In contrast, the intent in somatic cell nuclear transfer, or therapeutic cloning, is not to create a new person but instead to generate embryonic stem cells in vitro, in a culture dish, for customised cell therapy. It does not involve the creation of new life with a unique genetic combination, but is rather the propagation of already existing life, the life of a patient which we are trying to prolong. And the cloned embryo, as I mentioned before, has a very low potential, if any, to ever grow into a normal baby.

(Click on image for a larger version)

This leaves you with three potential fates for embryos derived either by fertilisation or by nuclear transfer.

The fertilised embryo can either be disposed of, as I assume many of those embryos are. It has a very high potential to give rise to normal babies, and it also has a very high potential to give rise to normal embryonic stem cells.

A cloned embryo, on the other hand, can also be disposed – many are in favour of that. The alternative would be to transplant the embryo into a female, where they would almost certainly grow into an abnormal baby. Most people object to that. However, the embryo has the same potential to grow into an embryonic stem cell line when put in culture. And these embryonic stem cell lines derived by nuclear transfer are not, by any means that we can test in the lab, different from embryonic stem cells derived from a fertilised embryo.

To conclude what I have just told you: we can say that reproductive cloning faces principal biological barriers – faulty reprogramming – that may not be solvable in the foreseeable future. Somatic cell nuclear transfer, on the other hand, does not face these principal barriers but only technical obstacles, which may be overcome in the near future.

When this is taken into account, the cloned embryo has two possibilities. It can be implanted in a womb, where it has very little if any potential to grow into a normal baby. Or it can be explanted in culture to derive embryonic stem cells, and these cloned embryonic stem cells can be used for custom-tailored therapy. They can also be used – and this is another very important use of this technology – to study complex human genetic diseases in a culture dish, because it allows you to, basically, transfer the disease from a patient right into a petri dish where you can then study the genes that are involved in disease progression. You can try, for example, to screen for drugs that would interfere with the development of the disease – all of the tests that you would not be able to do in the patient.

In my last few slides I would like to switch topics slightly and give you two examples of the use of nuclear transfer, not in cell therapy or for reproductive cloning but rather in basic science.

It was actually basic scientists, developmental biologists, who invented or discovered nuclear transfer as a tool to answer specific biological questions. It was in the 1950s and ’60s when Briggs and King, Gurdon and other researchers established nuclear transfer as a tool to study the nuclear potency of cells. This was all done in the amphibian system. They basically asked: are adult frog cells any different from embryonic frog cells? Is there a qualitative difference in the nucleus of those two cells?

(Click on image for a larger version)

What they observed, about 50 years ago, was that the efficiency of cloning frog embryos was much higher when taking early embryonic frog cells; it readily declined when cells from more committed stages were taken as donors for nuclear transfer.

This raised a historic cloning question: are fully committed cells still competent to give rise to an entire new animal, a cloned animal? This question was not fully resolved by the frog cloning experiments, and neither by the cloning of Dolly and other mammals because of the lack of appropriate genetic markers that would mark a differentiated cell.

(Click on image for a larger version)

So what we did was to produce monoclonal mice by nuclear transfer from terminally differentiated immune cells. The reason we took immune cells was that they carry a genetic marker for their differentiation state. The genetic markers are those irreversible DNA rearrangements of the antibody genes which basically are responsible for producing this vast repertoire of antibodies and T-cells in our body. (You see here a picture of the first cloned pup that we got.)

So what were the lessons to take home from this experiment? The most important lesson was that mature, fully differentiated cells are indeed still capable of giving rise to an entire animal. So terminally differentiated cells are still genetically equivalent to early embryonic cells.

What this experiment also showed us was that cloning actually allows us to detect very subtle genetic changes of a given donor cell and to analyse the functional consequences of these subtle changes in the context of a whole mouse, a cloned mouse.

(Click on image for a larger version)

The other experiment, where we tried to take advantage of this principle, was a test for the reversibility of cancer by using nuclear transfer. We know from the cloning of Dolly and the cloning of monoclonal mice from immune cells that the changes that occurred during development of an embryonic cell into an adult cell or a differentiated cell are reversible, because cloning can turn back an adult cell into an embryonic cell which can make a new animal.

So we wanted to know whether we could extrapolate this finding to cancer and ask whether the changes that occur during the transformation of a normal cell into a cancer cell are also reversible by nuclear transfer.

(Click on image for a larger version)

We used a couple of different cancer models. The one that worked best in our hands was a melanoma model. In the left-hand picture of the head of this newborn mouse pup you see stripes of green fluorescent protein, basically indicating the contribution of melanoma-derived reprogrammed cells to different tissues in this mouse. Here you see the contribution to skin tissue wherever there are green stripes. We saw contribution to many other tissues as well.

What this experiment told us was that the nucleus of a melanoma cell could be reprogrammed into a pluripotent embryonic cell which retained a very broad developmental potency and contributed to most, if not all, tissues in the cloned mouse. In consequence, it meant that the cancer genome was still responsive to these early embryonic cues in the oocyte.

The second conclusion was that the cloning of this melanoma cell allowed us to study the functional consequences of a single melanoma nucleus in the context of the whole mouse. We were very interested to see that these mice not only succumbed to melanomas but they also developed tumours in other tissues where this melanoma genome was present. For example, we saw tumours in muscle cells and in neural cells, which were never seen in the original mouse model.

(Click on image for a larger version)

My last slide summarises what I have shown in the last two experiments: nuclear transfer is a very powerful tool to study basic biological questions; it is a test for the reversibility of any cellular state, whether that is differentiation, cancer or any other cellular state that involves epigenetic changes; and it is a neat tool to study the potency of a nucleus in the context of an entire animal. I mentioned the monoclonal mice and the cancer mouse, but I did not have time to talk about mice that we produced from post-mitotic neurons.

(Click on image for a larger version)

Here is a list of the people who were involved in that.

Questions/discussion

Question – I didn’t follow why the problem with faulty reprogramming would not potentially be an issue with therapeutic cloning. Could you not also have a tissue develop, when you stimulate that tissue, in which the epigenetic resetting is not correct, and therefore that tissue may express some gene that it normally doesn’t, or aberrantly express tissue?

Konrad Hochedlinger – I don’t think that is an issue, because again the process of embryonic stem cell derivation already selects for functional cells. So cells where you have abnormal or faulty reprogramming would most likely be selected against. Secondly, the genes that are most often affected by faulty reprogramming are imprinted genes which are important for fetal development but do not play a role for mature, terminally differentiated cells which we would use for therapy.

Question – Does that mean that you can’t really study epigenetic errors? If that was part of the disease profile, would it be possible to study epigenetic abnormalities in embryonic stem cells?

Konrad Hochedlinger – It would be very difficult, because embryonic stem cells are very unstable epigenetically, in terms of imprinted genes and other genes. I guess you were referring to cancer, for example, where you want to look at tumour-suppressor genes or oncogenes which are aberrantly methylated. Yes, I think that might be a limitation for that approach.

Question – Is there any research being done on the importance of enzymes in resetting?

Konrad Hochedlinger – Our lab and many other labs throughout the world are working on that right now, trying to find out what are actually the enzymes that are responsible for these epigenetic modifications, and whether we can either overexpress them or eliminate them and then remove these abnormal epigenetic changes as a precondition for nuclear transfer.

Question – Does the oocyte that you transplant the nucleus into have to be from the same species?

Konrad Hochedlinger – I have never done trans-species experiments, but other people have done that. I believe they have used rabbit oocytes as recipients for human donor cells, so in certain combinations it works. But I think that in other combinations it doesn’t. There was a report that they had done all sorts of combinations, of cow into cow, rat into cow, which never worked, and human into rabbit, which apparently did to a certain degree.

Chair – I think you might have to reset the mitochondria as well, because you have got mitochondrial DNA and they are related to the nuclear messages. So you would probably have to reset the mitochondria – put in the same species as the nuclear donor.

Question – Obviously what you are doing is moving a differentiated nucleus, which you have probably got to strip in some way, in terms of its imprinting, back into the cytoplasm of an oocyte, and at some times that works better than at others. So how much do we understand about what the cytoplasm is contributing to that process?

Konrad Hochedlinger – We don’t really know much about it, actually – very little. There has been no reprogramming factor identified, or enzymatic activity that would explain how reprogramming works. People who are actively fishing for these factors are using candidate approaches to see whether methylation enzymes – DNA and histone methylating enzymes – may play a role, but nothing is really known about that.

Question – You have never actually taken the cytoplasm out of the oocyte and put it back into a somatic cell? I don’t know whether, technically, you could even do that.

Konrad Hochedlinger – I have never tried it. I think it would be technically very difficult. But one thing that has been done in Alan Trounson’s lab is people have tried to use embryonic stem cells as donors, or a source, for reprogramming factors, because when you fuse embryonic stem cells with a somatic cell you also get reprogramming, at least to a certain extent, of the somatic genome. So whatever is present in the oocyte also appears to be present in the embryonic stem cell. Many labs are trying to use embryonic stem cells now as a potential source to fish for reprogramming factors.

Chair – It seems that you have to have a nucleus at least for a period of time, so there must be transcription factors that are emanating from the nucleus and are responsible. You can’t simply do it by using cytoplasm.

Question – If the cytoplasm of an oocyte will restore totipotency to a somatic cell nucleus, how about the cytoplasm of a spermatocyte?

Konrad Hochedlinger – A spermatocyte doesn’t have much cytoplasm. I have never tried it. I know people have tried to use zygotes, for example, and earlier stages of oocyte maturation, as recipients for nuclear transfer and it did not work. So you have to have this particular stage, metaphase 2, of oocyte maturation as the recipient, for nuclear transfer to work.

Question – Once the cells have been added back and they differentiate into, say, liver cells or muscle cells, have you then looked at the epigenetic reprogramming, to be sure it is like normal cells? I just wondered if this fitted with what we heard this morning about the cord blood cells, that they seemed to take up the right patterns of gene expression if you put them into a tissue rather than just keeping them in tissue culture. The idea is that the tissue into which you put the ES cells is somehow signalling to them to take up the epigenetic modification of that organ. Is there some sort of signalling in to these cells to tell them what sort of epigenetic programming to have?

Konrad Hochedlinger – Before nuclear transfer, you mean?

Question (continued) – Any of the cells, actually, rather than the nucleus. Once you have the ES cell and then you put it back into, say, the mouse – as your skin cells went back into the mouse’s ES cells – and then, as in your case, they become part of the immune system (or, as we heard in a session this morning, the cord blood cells take up part of the tissue) then they become like the normal tissue. Is there some sort of signal from the tissue to those cells that you put in, to convert them?

Konrad Hochedlinger – I am sure there is. We didn’t really study that in detail. And, as I said, the restoration of the immune phenotype was only partial in this therapeutic cloning experiment.

Question (continued) – So you didn’t look at the epigenetic modifications of those cells?

Konrad Hochedlinger – We didn’t really look into that.

Chair – I guess they might also be selected for or against. There may be some selection advantage if they end up in a liver, or in a lung – and that is the way you make a chimaera. It is just possible that there is also selection there for the cells. I don’t think we have an answer at this stage, but there are bound to be some local cues which tell cells what they are meant to be doing.