Reprogrammed life

Interview with stem cell researcher Shinya Yamanaka.

SHINYA YAMANAKA, the Japanese stem cell researcher who is tipped to win the Nobel Prize, was the featured speaker in the third edition of the highly successful and popular Cell Press-TNQ India Distinguished Lecture Series (in biology) held in Bangalore, Chennai and New Delhi early this year. Yamanaka delivered a lecture on New Era of Medicine with iPS Cells based on his revolutionary research on induced pluripotency stem (iPS) cells, which began in 2006 and has brought closer to reality the promise of stem cells to provide cell-based therapy, in particular through what is called regenerative medicine' the process of replacing or regenerating human cells, tissues or organs to restore or establish normal function for various intractable and otherwise incurable health disorders.

Yamanaka's breakthrough work lay in the creation of embryonic-like stem cells from human somatic cells mature biological cells from different parts of the body such as skin cells, blood cells and muscle cells. In the words of another well-known stem cell scientist, Robert Lanza of Advanced Cell Technology and Adjunct Professor at the Institute for Regenerative Medicine at Wake Forest University, [Yamanaka's work] is likely to be the most important stem cell breakthrough of all time. The ability to generate an unlimited supply of patient-specific stem cells will revolutionise the future of medicine (emphasis added). This work has opened up an entirely new avenue of research on stem cell, and many laboratories around the world are now engaged in this field of iPS cells. It has been reported that following the success of experiments spawned by his work, which obviates ethically problematic use of embryos in biological research, Ian Wilmut, who helped create the first cloned animal, Dolly the Sheep (July 1996), has announced that he will abandon nuclear transfer as an avenue of research! His work was voted as the Method of the Year in 2009 by Nature Methods.

Yamanaka is at present the Director of the Centre for iPS Cell Research and Application and Professor at the Institute for Frontier Medical Sciences at Kyoto University. He is also a Professor of Anatomy at the University of California, San Francisco (UCSF), and a Senior Investigator at the UCSF-affiliated J. Gladstone Institute of Cardiovascular Disease. During his distinguished career in science, he has been honoured with several awards, including the Robert Koch Prize (2008), the Albert Lasker Award for Basic Medical Research (2009), the March of Dimes Prize (2010) and the Wolf Prize in Medicine (2011).

Interestingly, Yamanaka started out his career as an orthopaedic surgeon but found that he was not so good at surgery. He felt that as a surgeon he was not able to help many patients. Also, he says, very, very good surgeons cannot help patients with intractable diseases and injuries, such as the motor neuron disease amyotrophic lateral sclerosis (ALS), or Lou Gehrig disease as it is popularly known. So I decided to move to basic medical sciences. During his days as a PhD student pursuing research in pharmacology, he got interested in knockout (KO) mouse technology. In order to learn more about KO technology, he went to the United States in the mid-1990s to be a postdoctoral fellow at the Gladstone Institute. That is where he came across embryonic stem (ES) cells because mouse ES cells were needed to make KO mice. During the course of his research, he had identified a new gene that seemed to have significance for cancer. In order to study that gene, he made KO mouse and discovered that the gene was important for pluripotency (the ability to differentiate into over 200 types of cells in a mammalian system) in ES cells.

Two events happened in the 1990s, he says. One was what he calls PAD (post-American depression), which afflicted him in 1998 on his return to Japan when he was frustrated that he could not continue his work in the U.S., and two, setting up his own laboratory in 1999 at the Nara Institute of Science and Technology at Nara, Japan, where he had joined on his return. It was then that he set a clear, long-term goal of reprogramming adult human cells to ES-like cells.

With the demonstrated potential of iPS cells to be transformed into hepatocytes, neurons, cardiac muscles and pancreatic beta cells, there is clear evidence today that scientists are close to achieving iPS cell-based directed cell transplantation therapy for spinal cord injuries; diseases such as Parkinson's, ALS, muscular dystrophy, Type I diabetes; liver diseases such as cirrhosis; and eye disorders such as age-related macular degeneration (AMD). Most importantly, iPS cell-based therapy is devoid of ethical issues involving ES cells and potential immune rejection with ES cells because here iPS cells can be patient-specific by generating them from the patient's own somatic cells. Only on April 18, a group led by Deepak Srivastava, a scientist of Indian origin at the Gladstone Institute, using the method developed by Yamanaka succeeded in converting scar-forming cardiac cells into beating heart muscle by direct injection of iPS cells into living mice. This had been previously achieved by scientists in Petri dishes. Given this important therapeutic potential of iPS cells, Yamanaka seems to wish to go back to the clinic and treat patients when he says, Now I want patients for receiving iPS cells.

Yamanaka spoke at length to Frontline's Science Correspondent R. Ramachandran during his Cell Press-TNQ lecture tour. But before we reproduce the excerpts from the interview, a brief introduction to this highly technical subject of iPS cells would seem to be in order.

Stem cells, ES cells and iPS cells

Stem cells, as we know, have the remarkable property to develop into different cell types in the body during early life and growth. Further, stem cells act as a sort of repair machinery, dividing essentially indefinitely to replenish other cells for the lifetime of the person or animal. Stem cells have two important properties that distinguish them from other cell types such as muscle cells, skin cells, blood cells or nerve cells. 1) They are unspecialised, capable of renewing themselves through cell division over long periods (proliferation) and 2) they can give rise to specialised cell types. Stem cells give rise to specialised cells through the process called differentiation.

Until Yamanaka succeeded in creating iPS cells, scientists worked with two kinds of stem cells: ES cells and somatic or adult stem cells. ES cells are derived from embryos resulting from in vitro fertilisation of eggs from the inner cell mass of the blastocyst, an early stage embryo (4-5 days old) when they consist of about 50-150 cells. The chief characteristic of ES cells, besides their ability to proliferate indefinitely, is their pluripotency, that is, the ability to differentiate into all types of cells in the adult body. Adult stem cells, unlike ES cells, can typically produce only the cell types of the tissue in which they reside. For instance, stem cells in the bone marrow can give rise to many types of blood cells. Adult stem cells predominantly serve to maintain and repair the tissues of the niche environment in which they are located. Though many experiments have reported that certain adult stem cell types can differentiate into cell types other than those of their normal niche (transdifferentiation), it is not clear whether this phenomenon actually takes place within the human system. In a variation of these experiments, scientists have also demonstrated that certain adult cell types can be reprogrammed into other cell types in vivo using a well-controlled process of genetic modification or transfection, say connective tissue cells (fibroblasts) to muscle cells (myocytes).

Enter Yamanaka. What he showed was that as against reprogramming the cells to become only a specific cell type, it was now possible to transform adult somatic cells (not somatic stem cells) to become like ES cells through the transfection of certain embryonic genes with identical properties of proliferation and pluripotency. Hence, the name induced pluripotent stem cells. The adult cells are forced to express genes and transcription factors (through over expression) important for maintaining the chief characteristics of ES cells. Although iPS cells have been found to be equivalent with regard to all the properties to ES cells in vitro as well as in vivo in experiments with mice, it is not clear whether iPS cells and ES cells will differ in their clinical application in humans.

Yamanaka reported mouse iPS cells in 2006 and human iPS cells in 2007. Early observations that somatic cells could be reprogrammed either by nuclear transfer into immature egg cells (oocytes) or by fusion with ES cells suggested that these contain factors that can induce reprogramming. Yamanaka identified these factors and conceived that it should be possible to induce pluripotency in somatic cells without the use of egg cells or embryos.

iPS cells thus derived have been found to display the important pluripotency characteristics of ES cells, including the formation of teratomas containing cells from all three germ layers (endoderm, mesoderm and ectoderm), and stem cell markers and being able to differentiate into different tissues when injected into mouse embryos at very early stages of development, including the formation of chimeric mice (mice exhibiting different genetic make up in patches derived from two or more embryos by experimental intervention).

Reprogramming to generate iPS cells poses several issues before they can be clinically used for regenerative medicine. One of the important issues that arises from the mechanism used for transfecting the transcription factors is that the DNA (deoxyribonucleic acid) which encodes for their production must be introduced and integrated into the genome of target somatic cells. This was originally achieved by using retroviral vectors. A retrovirus is an RNA (ribonucleic acid) virus that uses the enzyme reverse transcriptase to replicate in the host cell and subsequently produce DNA from its RNA genome. This DNA incorporates in the host genome, allowing the virus to replicate as part of the target cell's DNA. But such forced expression of genes using retroviral vectors has been found to lead to unpredictable effects. Similarly, one of the transfected factors is potentially cancer-causing. However, recent research using non-integrating adenoviruses or transient plasmids and episomal vectors has shown that genomic integration itself is not necessary for reprogramming to occur. As Yamanaka says in the interview, We are closer to overcoming these major hurdles.

Excerpts:

Professor Yamanaka, when you started looking at stem cells what was it that indicated to you that generating embryonic-like stem cells from somatic cells was possible?

First of all it was the birth of Dolly, the sheep, which Dr Ian Wilmut generated from somatic cells by nuclear cloning. From his experiment we knew that it should be possible to reprogramme somatic cells back into the embryonic state. That was one thing.

Another thing was that approximately 10 years ago Dr Weintraub showed that by introducing only one transcription factor, which is MyoD, into fibroblasts [the principal active cells of connective tissues] it was possible to convert fibroblasts into myocytes just by one transcription factor. It was in mice. Also in the fly, Drosophila, you can convert eye into leg and antenna into leg just by introducing one transcription factor. So, from their success we knew that transcription factors can convert cell fate from one lineage into another lineage. From the combination of those two precedent results, I thought that by either introducing one factor or a combination of several factors we might be able to convert somatic cells into ES-like cells.

Stem cells are characterised by not just one transcription factor but by many transcription factors and by many surface proteins

Exactly

while these previous experiments and your later experiments suggest that either one or a few transcription factors could be sufficient How do we know that the iPS cells will function exactly like ES cells and will have exactly the same properties?

It is quite remarkable that iPS cells are nearly indistinguishable from ES cells. They used to be just skin cells or blood cells. But they are now really just like ES cells. They are not just similar. They are almost identical. I think the reason for that is that ES cells are so stable. It is like a hole in the ground. If you drop a ball on the ground, the ball starts rolling but it will stop at the bottom of the hole. So, that is the state of the ES cells and also that is the state of the iPS cells.

Nevertheless, the question of strict equivalence between ES cells and iPS cells is something which seems to be not settled. Are there any key differences that you see between the two say, in terms of their proliferation and differentiation characteristics?

Actually, no. Some scientists reported when they found some differences. However, we could not find such differences. If you compare only a handful four or five of iPS cell lines and ES cell lines, you may find some differences. But if you compare 30 or 50 cell lines, those differences are not consistent. It's just like comparing people. If you compare, say, five people here and five people in other places, you may find some differences. But if you compare more people, you cannot find such differences. So each cell line has some variation. So if you compare small number of cell lines, you may claim Well, we have identified some differences. You just have to study more cell lines.

You created these first iPS cells using four transcription factors. Do they suggest that these are an indispensable set of genes or could there be a reduced set or even a different set of genes which could do the job?

I think so. I don't think that the combination of those four factors is the only combination. Other combinations may be able to do the same thing. As a matter of fact, James Thomson [developmental biologist] in [the University of] Wisconsin reported their generation of iPS cells with four factors, but only two are common with ours. They have two unique factors; we have two unique factors. From this one story, we can say that our four factors are not the only combination. But the efficiency is still very low. So that means that probably we still need some very important factor.

But do you already see that one of them seems to be indispensable?

Probably, the factor Oct4 is indispensable. But other factors like Sox2We have already shown that Sox2 can be replaced by others from the family of Sox transcription factors. Similarly, c-Myc can be replaced by l-Myc and n-Myc, and even other factors; Klf4 can be replaced by Klf2, and even by Klf5. So they are not indispensable, but I think Oct4 is indispensable.

But the transcription factor c-Myc is supposed to have carcinogenic properties. How about l-Myc and n-Myc? Are these also carcinogenic?

l-Myc has very, very weak oncogenic [carcinogenic] activity. So it is different from c-Myc or n-Myc. However, we found that l-Myc is more potent in terms of iPS cell generation. So, I think we can distinguish iPS cell generation from transformation from that experiment. We now use l-Myc to make human iPS cells for the purpose of regenerative medicine.

You mentioned in your talk that if you used six factors, the efficiency was more than using only four. Does that tell us something?

We demonstrated that introducing four genes into somatic cells can induce pluripotency, but this is just one of the ways to reprogramme cells. We actually showed a three-factor method to generate iPS cells from somatic cells. There is a possibility that a method of combinations of six factors is more efficient than a method using four factors.

But do you see any significant variability in the characteristics of these iPS cells perhaps not so much in terms of proliferation and differentiation when you use different sets of factors?

That's kind of difficult to answer...

But is there variability in the different lines that you produce?

Yes. Yes. What is important to realise is that reprogramming can be full or can also be only partial. Partial [iPS cell] lines are still very similar to ES lines. But they have many differences from ES cell lines and completely reprogrammed iPS cell lines.

So, in the experiment that you are doing, how do you know that reprogramming has been achieved completely?

That's a good question. That's what we have been working on very hard for the last four-five years since we reported our first reprogrammed human iPS cells. By morphology alone we cannot tell. Some big bad iPS cells we can tell by their morphology itself. But many partially reprogrammed iPS cells look okay. However, when we looked at gene expression patterns, they have some different expressions; they have some different DNA methylation [biochemical process important for normal development] patterns. Now, we can distinguish those partially reprogrammed iPS cells from fully reprogrammed iPS cells.

So what are the key tests that you do to determine this?

I think if you do DNA microarray analysis [technique used to measure expression levels of genes], you can just distinguish.

But in terms of its gross properties, one of the tests that you do is to look for teratoma formation using iPS cells. Is that alone not sufficient to say that iPS cells would function like ES cells?

No. Not at all. Even partially reprogrammed iPS cells can form teratomas. So you can't tell from this that this iPS cell is very good just by teratoma formation. So in humans this is a very big problem. In mouse we have chimera assay. We microinject iPS cells in mouse blastocyst. If iPS cells are very good, you can obtain chimeric mice. By crossing chimeric mice with other mice, if the iPS cells are very good, you can obtain the so-called germ-line transmission [the process of passing on germ cells containing the genetic material to the offspring]. That's a very stringent test for mouse iPS cells. However, we cannot perform that kind of test with human iPS cells. Teratoma formation right now is the test for human iPS cells. Many iPS cells can make teratomas but they cannot make chimeras. So we really have to distinguish the good ones from the bad ones. As I mentioned, we do that by DNA micro-array analysis and DNA methylation analysis.

ES cells can remain in an undifferentiated state for a very long time and once they differentiate, the differentiation is sustained. How to ensure the same thing for iPS cells? That is, the differentiation occurs at the right time and is also perpetual?

Well, It's very important. In other words, you are asking how to maintain the undifferentiated state for a sufficiently long time. We really have to have good culture conditions to achieve that. But thanks to more than 10 years of many researchers using human ES cells, we now know how to maintain ES cells and we can use the same conditions for iPS cells. The difference between the two is only the origin. Everything else is almost identical. So we can use the many results of researches with ES cells and translate them into iPS cells.

And also how to ensure that once differentiated, this differentiation is unique and directed and it does not result in other kinds of differentiation?

Yes. It is very important. If you want to treat a patient with Parkinson's disease, you really have to make dopaminergic neurons from iPS cells or ES cells and then transplant them into the patient's brain. [Dopaminergic neurons of the mid-brain are the main source of dopamine in the central nervous system. Their loss is associated with Parkinson's.] We do not want to transplant anything else. With any of the current procedures that we have right now, we cannot make one hundred per cent dopaminergic neurons with ES or iPS cells. We always end up with something else in addition to dopaminergic neurons. So what is very important is how to purify dopaminergic neurons from those contaminating differentiated cells of other types and other persisting undifferentiated [ES or iPS] cells. That's a very important issue in this field. We could do that by flow cytometry techniques if we could identify a good surface marker. So that's what many researchers have been working on, to find unique surface markers. So that's the key issue. For many types of cells we can now do that.