Martin Pera is Research Professor at the Monash Institute of Reproduction and Development at Monash University and the Director of Embryonic Stem Cell Research at the Australian National Stem Cell Centre. He carried out postdoctoral research at the Institute of Cancer Research and the Imperial Cancer Research Fund in London, and was a Research Fellow at the Department of Zoology at Oxford University. His research interests include the cell biology of human pluripotent stem cells, early human development, and germ cell tumours. He was among a small number of researchers who pioneered the isolation and characterisation of pluripotent stem cells from human germ cell tumours of the testis, which provided an important framework for the development of human embryonic stem cells. His research group currently focuses on the extrinsic factors involved in maintenance of the pluripotent state in human embryonic stem cells, and those factors that drive their commitment into progenitor cells representative of the three embryonic germ layers.

Human embryonic stem cells – the state of the art
by Professor Martin Pera

I am delighted to have the opportunity to speak in this prestigious forum, and to set the scene for today’s talks. Because this is a general audience, I am going to begin by defining some terms and laying some groundwork that will, hopefully, make the job of the later speakers a bit easier. Then I am going to give you an update on the current state of research in my own field, human embryonic stem cells.

Definitions

Let’s start off by defining what we mean by a stem cell. A stem cell is a primitive cell – primitive, in terms of having few distinguishing morphological features – with two key properties. The first property is the ability to undergo self-renewal, that is, to divide time and time again to produce more stem cells. The second property is the ability to undergo differentiation or specialisation to give rise to mature functional cells. These properties mean that stem cells have the potential to replace dead or damaged cells in diseased tissue.

I am going to talk about differentiation today. Differentiation is a result of changes in gene expression. That is not all it is; it is a complex process whereby a cell acquires the right shape, polarity, orientation with respect to neighbouring cells, the appropriate internal apparatus – organelles – and the proteins that enable it to do a specific job in a specific tissue.

Types of stem cell

We are going to talk today about a couple of different kinds of stem cells. One type of stem cell we are going to talk about is sometimes called ‘adult stem cells’. I don’t particularly like that term, for reasons I will explain. Instead I use the term ‘tissue stem cells’. These are stem cells that reside in established organs or tissues, established after the period of embryogenesis is complete – in other words, after the first trimester of development in man. These include not only ‘adult’ stem cells, if you will, but stem cell populations of foetal, neonatal, paediatric and adult tissues. In general, these stem cells are committed to form a limited range of cell types. We say they are multipotent, if they can form several cell types; bipotent, if they make two cell types; or unipotent, if they perhaps make only one.

Tissue stem cells exist in adult or foetal populations, in skin, in the hair follicles, in the lining of the gut, in the brain and in the blood cells. It turns out that the stem cells in these tissues live in a very particular part of the tissue, a particular microenvironment or niche. The study of these microenvironments or niches is a very important area in our field today, because if we understand what the environment is and how it interacts with the stem cells, we will be better able to control and propagate stem cells outside of the body.

Recent discoveries within the past decade have indicated strongly that tissue stem cells have a lot of untapped potential for repair and regeneration. It was originally thought, prior to this more recent era, that stem cells only resided in tissues that underwent constant renewal, those tissues where the mature cells are lost on a regular basis and there is need for continual replacement, like the skin, the hair follicles, and the lining of the gut.

More recently, stem cells have been discovered in tissues that were long thought to be quiescent in adult life, tissues like the brain, the heart – these are fascinating new discoveries. What those stem cells do there in normal physiology, and what their potential to repair tissue damaged by disease, are very exciting and rapidly progressing areas of research.

Finally, there are what I call facultative stem cells in certain tissues. These are stem cells that are quiescent most of the time, that do not normally have a role in renewal and may not even function in all types of repair, but that act as a sort of reserve. They can be called into action in certain types of damage, and play roles in particular types of tissue repair. There are probably cells in liver and pancreas, for instance, that fit this category.

There are some properties of tissue stem cells that pose challenges to their use in research and therapy. They are generally very rare, a small fraction of the total population of a tissue. This means it is sometimes difficult to isolate them in pure form. And, although some can be grown to a degree outside of the body, they generally have a limited lifespan outside of the body and many can’t be propagated at all.

In recent years we have heard a lot about tissue stem cell plasticity. What is plasticity? It was long held that tissue cells are committed to a particular fate, to produce only a limited range of cell types. But more recent studies have indicated that tissue stem cells are probably much more flexible than we previously appreciated, and that they can even cross developmental boundaries.

Our concepts concerning stem cell plasticity are in fact in a state of flux themselves. In 2000 and 2001 there were many surprising reports of tissue stem cell plasticity. Brain was shown to make blood, blood to make brain, and many other examples were reported in major journals. A few years later many of these studies were subsequently challenged in further work.’

Some of the studies have proven difficult to repeat, or perhaps alternative explanations for the findings have emerged. And these transitions of tissue stem cells to make other cell types often occur at low frequency and only in response to severe tissue damage or an altered environment.

There are very good reasons for strict control over tissue stem cell fate. Those controls are there to ensure proper regeneration and repair of tissues, to make sure we don’t suddenly get bone growing where our cornea should be. These are limits that are strictly imposed by powerful restraints on gene expression, and they are heritable through many rounds of cell division. Nevertheless, it is quite clear that under certain circumstances stem cells can show relaxation of these restrictions. Understanding what the conditions are that drive that relaxation is important, because it may one day enable us to unlock the potential of tissue stem cells.

Embryonic stem cells

These are derived in the human from five- to seven-day-old spare human embryos, before the specialised tissues of the body have begun to form. They have two key properties: embryonic stem cells can be propagated more or less indefinitely in culture in the primitive embryonic state; and while they are growing they retain the property of what we call pluripotency – the ability to give rise, not to a few but to all the cell types of the adult body. And they retain this property during extended growth in vitro. Those two features, the ability to propagate indefinitely and the pluripotentiality, are what give rise to all the excitement about human embryonic stem cells.

Human development, begins with the fertilised egg which undergoes cleavage divisions to yield two cells, four cells, eight cells. By about five days, we have the structure on the known as the blastocyst. The cells on the outside of that shell-type structure have already made a commitment decision: they are now restricted and they will only become part of the placenta. The clump of cells sitting on the inside will give rise to all the tissues of the body. That is the inner cell mass, and that is what we make embryonic stem cells from.

the properties of pluripotent stem cells are as follows. They originate from cell populations that are ‘naturally’ pluripotent, such as the inner cell mass. They can be propagated indefinitely in vitro. While they are growing they maintain a normal genetic make-up. A single one of these cells is capable of differentiation into a wide range of tissues, in vivo and in vitro, at high frequency and under a range of conditions.

In the mouse a very powerful experiment shows the capabilities of stem cells, and this is to take the embryonic stem cells, reintroduce them into a host embryo, take that host embryo and put it into a foster mother, and let it develop to term. What emerges is a chimaeric animal. If you do the experiment in a certain way, you can make all the tissues of that mouse derive from embryonic stem cells. This experiment shows not only that you can make all the tissues of the body but that those tissues can function throughout the life of the animal in a normal way.

Before I leave this topic of embryonic and tissue stem cells, I would like to point out that research in these areas is complementary and synergistic. To date, no-one has isolated a tissue stem cell that really has the properties of an embryonic stem cell.

Our Australian Stem Cell Centre, of course, funds both types of research. And it is very important to understand that this is new science. It is way too early in our studies to assert that one of these avenues or the other is going to be ‘the’ way to go for a particular clinical application. I think to formulate policy on the basis of what are essentially preliminary findings in a new, emerging field is foolhardy.

Embryonic stem cells have important applications in biomedical research: in basic studies of early human development and its disorders; in the discovery of new factors that control tissue regeneration and repair; potentially to provide in vitro models for drug discovery and toxicology; and – the application that has been most emphasised – with a role as an indefinitely renewable source of tissues for use in transplantation medicine.

Embryonic stem cell research today

Since the first report on embryonic stem cells, the growth in publication in this area has been exponential.

There are challenges facing us, however: learning how to better propagate the stem cells and scale up their cultures, understanding the stem cell populations, directing their differentiation efficiently down particular pathways, and translational research and delivery of embryonic stem cell based therapies.

In terms of the derivation of new stem cell lines, their characterisation and propagation, what we are after is better stem cell lines, a better understanding of what a pluripotent stem cell really is, and better means to grow the cells in culture.

We have heard about the need for new embryonic stem cell lines. Why is this? There are several reasons. Most of the current lines are unsuitable for therapeutic use because they have been derived in the presence of animal products, with a potential for disease or pathogen transmission across species. Researchers are devising better, more defined culture systems free of animal products, and we want to implement these in the second and third generations of cell line production. But there are still questions of open access. Many of the existing lines have restrictions on their use and distribution, and we want to get over that.

There are new derivations going on under legislation here and elsewhere, and it is important to remember, with regard to regulation, that ES cell derivation has absolutely no impact whatsoever on the production or disposal of human embryos in Australia. That will go on anyway; it is just a question of whether we discard them or whether we use them in a more constructive fashion.

Characterising the stem cell populations is an important activity as well, because we need to know how the properties of different stem cell lines – isolates – compare when they are grown under similar conditions. And we need to know how, when we change the growth conditions (hopefully, to improve them) these changes affect the properties of the stem cells. I would point out that if there are stem cell lines with particularly desirable properties, for example, a special ability to give rise to islet cells for transplantation for diabetes, and if they are rare – let’s say they are only present in 10 per cent of the population – that means we would probably require systematic evaluation of several hundred cell isolates, to make sure we find them and identify them.

Finally, we need criteria, as we move in a regulatory sense towards ultimate therapeutic use, to define what is a human embryonic stem cell culture.

We go at this in several ways. With the modern DNA technology and microarray technology, many workers have undertaken transcriptional analysis of stem cell populations. What this has done is to identify a core set of genes that are commonly found in human and mouse embryonic stem cells. The importance of this is that we can now have a molecular blueprint of the pluripotent state, and we can begin to understand how the components interact. That means we can understand how to control pluripotentiality, whether that is to maintain a stem cell or to reset the developmental machinery of an adult cell.

Additionally, we use immunological markers to characterise stem cell populations.

And we can use modern technologies, such as antibodies and flow cytometry, to isolate sub-populations of stem cells.

What this work in my lab and others is telling us is that human embryonic stem cell cultures are actually complex cell societies, that the differentiation process begins probably earlier than we had ever conceived, and that interaction between the stem cells and these early differentiated cells strongly affects cell fate.

In the mammalian embryo shortly after implantation, the key decisions are taken to specify the body plan – what part is going to become brain, what is going to become precursors of the skeleton, and what have you. Those decision events, we know now, are mediated by interactions between the pluripotent cells, and the surrounding so-called extra embryonic tissues.

Similar conversations probably occur in our stem cell cultures, and we are beginning to identify what some of those conversations are. If we know the players, we can regulate those interactions and get desired, directed differentiation, instead of complex mixtures of cells.

When it comes to culture, we need to devise new techniques to scale up the production of stem cells and to remove all the undefined animal products.

For example, a particular proprietary product, the most widely used additive for human stem cell cultures. It is a concoction of many types of animal protein.

So the current culture environment is complex. We have undefined animal additives, we have the feeder cell layer. We want to get to a state where we have a synthetic extracellular matrix, defined growth factors and defined low molecular weight components. So around the world there are many groups working on this, there are many new systems that have been reported. But really I don’t think anyone has really put it all together properly. I think most require extensive additional evaluation.

The key issues are scale-up and purity; elimination of the feeder cells and animal protein requirement; better growth from single cells (an important feature for many experimental manipulations); and genetic stability, making sure that all the time we are propagating the cells they retain a normal genetic make-up.

What we are moving towards is fully defined systems, firstly to get rid of the animal products but also because a fully defined system enables you to control the differentiation much better. The system will have to be built up incrementally, and we will have to do careful testing to make sure we get maintenance of normal stem cell phenotype and a normal genetic make-up. And there are international collaborative efforts, in which Australia is a participant through the Stem Cell Centre and the NHMRC, to begin by comparing all the cell lines that have been made, around the world, to achieve standardisation, and, hopefully, in future to look in an unbiased, interlaboratory comparison at these Cell Initiative.

Stem cell differentiaton

I would like to turn now to differentiation. The differentiation capabilities of human embryonic stem cells are impressive, shown by simply injecting the cells into a mouse with no immune system: the result is a benign tumour containing complex mixtures of human tissue, including muscle, gut, nerve tissue, skin et cetera.

We want to learn what the signals are that control the early commitment of stem cells, how we can expand precursors committed to particular fates to yield large numbers of mature cells, desired types, in pure form.

We take our cue from studies of the mammalian embryo, which have given us some idea of the major molecular players in those fate decisions.

I will give you a quick time scope over where we are in some of this work. In November 1998, Jamie Thomson published the first derivation of embryonic stem cells.

Two years later, Ben Rubinoff, in our labs, confirmed Thomson’s work and showed further that human stem cells could form neural tissue in a culture dish.

A year later, Ben Rubinoff, again working with us, showed that you could isolate neural precursors from human embryonic stem cells and propagate them. And he characterised these neural precursors. That was also confirmed at the same time by Su-Chun Zhang, in Jamie Thomson’s lab.

By 2004, we and others had worked out ways of driving stem cell differentiation in this particular direction. We used an embryonic protein called noggin. What we were able to show was that treatment with this molecule induces efficient differentiation of stem cells into these primitive neural precursors.

Later in 2004, Reubinoff, using our techniques in Israel, showed that these noggin-derived neural progenitor cells could provide improvement of function in a rat model of Parkinson’s disease.

In about the same time frame, Norio Nakatsuji and his colleagues in Japan, using monkey embryonic stem cells, used a primate model of Parkinson’s disease and showed that beneficial improvement in symptomology and in some of the biochemical parameters could be achieved by grafting.

In another line of work, just this year, a couple of groups have demonstrated that cardiac muscle cells derived from human embryonic stem cells can restore pacemaker function in a couple of models of heart block.

There is a wide list now of cell types that have been derived from human embryonic stem cells in vitro. While the list is impressive, but there are some real caveats.

We want to induce the differentiation reproducibly, in a controlled, stepwise fashion. We want to make sure the majority of the population responds. We want to understand those factors that are controlling the process. We want to be able to identify, propagate and expand progenitor cells at various stages along the line. And we want to make sure we are getting differentiated cells with the expected patterns of gene expression and, importantly, functional capability. In very few of those instances have we come near achieving all of this, so there is a long way to go.

Nevertheless, there has been successful treatment of a number of animal models of disease with mouse embryonic stem cell lines.

I think the real challenge is still out there. For many cell types we still have to produce the required cell type in sufficient numbers and pure form. We have to think about what cell to transplant – do we transplant a relatively immature neural precursor to cure Parkinson’s disease, or do we want the mature dopaminergic neuron, which is the cell type that is actually missing? There are questions of how to deliver the therapy – in some cases this is simple, in other cases it is complex. There are problems of tissue rejection.

And a very, very interesting question here that is emerging from work both in adult and embryonic stem cells is: what are those grafted cells actually doing? Are they really just replacing dead cells that are missing, or are they interacting with the environment in a complex way so as to promote protection of the endogenous cells, or to stimulate endogenous repair?

The last application is that stem cells are discovery tools. We have now a renewable source of human, normal, diploid cells that we can study in the laboratory. And there are a number of applications for this – for instance, in functional genomics. Other applications in research exist. Pete Schultz and Sheng Ding, at the Scripps Research Institute, are chemical biologists who are using stem cell based high-throughput assays to screen chemical libraries for small molecules that affect cell differentiation, or cell commitment. This is a powerful approach, because it gives us new tools to look at the pathways that are involved in cell commitment; it also gives us lead compounds for pharmaceuticals that might one day be used to influence tissue regeneration or repair.

Nuclear transfer

I am going to finish up by saying a few words about nuclear transfer.

Somatic cell nuclear transfer begins with the removal of a cell from an individual. The nuclear material of that cell is placed into an egg that has had its own genetic material removed, and that egg is stimulated to begin development. When the cloned embryo gets to the blastocyst stage, an embryonic stem cell line is produced, from which you can produce tissue that is – with the exception of its mitochondrial DNA – genetically matched to the original patient.

Nuclear transfer to make stem cell lines combines cloning methodology with embryonic stem cell technology to produce these cells which, as I said, contain the genome of an existing individual. It was originally proposed mainly as a promising solution to the potential problem of tissue rejection. In other words, existing embryonic stem cells are foreign to your tissue and there is a chance your immune system might reject them.

This so-called therapeutic cloning – has been proposed but we can ask whether it is necessary or feasible. We don’t yet know how severe the problem of immune rejection of these grafts will be. There are certain aspects of embryonic-derived cells that appear to make them less visible to the immune system;

We also have to ask whether this process would be practical in the clinic. Where will the eggs come from, and can the procedure be turned around in the required time frame? And is it safe? It does appear that it is a little bit easier to make normal cells from cloned embryonic stem cells than it is to make an entire animal through cloning, which is a very inefficient process. Nevertheless, we still face the question: if we make a line from an individual patient, what are going to be our safety criteria?

However, I think this technology does have very, very important applications in research that justify our exploring it. First of all, it might be an easier way to produce banks of stem cells with particular tissue-matching characteristics. That is, if you had a bank of stem cells with a particular set of tissue types, you could match a significant degree of the population, and it may well be easier to do this in a directed fashion using nuclear transfer than to do it in a haphazard fashion through IVF.

Secondly, this approach would enable us to produce cellular models of specific complex human diseases. We could take cells from affected individuals, make cell lines, differentiate the cell lines into the tissues whose pathology is affected, and study the disease process.

This also enables us to study the genetic basis of many common human diseases. Many common diseases have a genetic basis but it is multigenic – it is not just one gene. Understanding these disorders represent challenges even in the modern genomic era. Stem cells and genetic manipulation of stem cells in the laboratory setting may give us some insight into this.

Most importantly, I think this research can provide us with some understanding of how an adult human genome can be reprogrammed to the pluripotent state. Somatic cell nuclear transfer is the most dramatic example of reprogramming of adult cells, and I think if we could use this technology to understand this process, we might be able to understand how to exploit adult stem cells in a much more effective fashion.

Let me summarise where we are today. The original findings on human embryonic stem cells have proven robust and highly reproducible – pluripotent cells are readily isolated from the human preimplantation embryo. The technology is now widely disseminated. There have been some improvements to the culture systems and to our understanding of pluripotent stem cell phenotype. And there has been some progress in controlling differentiation and in demonstrating the potential function of human embryonic stem cells in tissue repair. But there is an awfully long way to go.

Questions/discussion

Question – I am interested in what the metabolic state of these cells is. There appears not to be a lot of attention given to maintaining particular partial pressures of oxygen and CO₂. Were they anaerobic metabolisers, et cetera? Can you say a bit about that, please?

Martin Pera – I think you have raised an important issue that really has been a bit ignored. I think the cell biologists have tended to focus on the growth factors and those signalling pathways. The more physical environment of these cells is of great importance, and I think to a degree it has been overlooked.

We have been using, for cancer cell lines, media that were established 30 or 40 years ago. There is certainly room to optimise that. And it is interesting: I think that some of the transcriptional analysis will point to biochemical pathways that are prominent in these cells and I think we will learn something from that, to address the questions. But it really has been ignored.

Question – Firstly, you raised the issue of heterogeneity in the embryonic stem cell colonies. Throughout the world we have more than 120 different lines available so far, and all are heterogeneous in terms of the number of cells present in these colonies. From 1998, when Jamie Thomson used these cell lines for the first time, up to today, the emphasis has been to produce these cell lines. But now you raise issues like the purity of the starting material to get the tissue. Where does the Australian Stem Cell Centre stand in terms of quality control of the starting material? And, since we have a limited number of clones available throughout the world, our lab is the first one in the world to produce these clones using FACS sorting. Where does the Australian Stem Cell Centre stand in terms of putting some more resources into quality control of the stem cell line in the first place? That is my first question.

Secondly, very quickly, I want to raise the issue of nuclear transfer. Australia has made big strides in terms of stem cell research, and we want to continue that lead. Do you think that somatic cell nuclear transfer will be an issue, or will be put up for discussion in the parliamentary legislation which is up for review this year?

Martin Pera – In regard to heterogeneity and quality control, we think this is very, very important indeed. The Stem Cell Centre has a number of programs that are addressing this. We are part of the international initiative which is collecting these many cell lines from around the world and characterising them. We are also active in developing new antibodies to fractionate the population, studies of gene expression, et cetera.

In terms of new derivation, I don't know how much we can control the original starting material, in terms of the embryo. But I think control of the heterogeneity will come when we understand the growth requirements better. I think a lot of the heterogeneity comes from differentiation that is ongoing because we haven't quite got the culture system right. So I think that is a very important aspect as well, and it is one which we are pursuing. The point you raise is indeed a very important one.

My answer to the second question is yes, I think it is really time to revisit this issue of somatic cell nuclear transfer. It is a contentious issue. I think our duty will be to educate the public, not that this is necessarily a panacea for all the transplantation problems but that it is a serious and powerful research tool for looking at functional genetics in the human, for looking at human disease states. It is up to us to get that message across, and that is what we have to do.

Question –  My question concerns one of your early slides. It was a surprise to me that there is mention of follicular tissue. Speaking as a 'follicularly challenged' individual, couldn't research in that area and possible success in therapeutic cloning finance the whole of the rest of the enterprise?

Martin Pera – Ah, no, sorry!

Question (continued) – The thing is that one of your early slides showed follicular stem cells.

Martin Pera – Yes. It is a controversial area.

Chair – It is terrible to remember that the two great drug successes of the 1990s were minoxidil for baldness, and Viagra.