Featured Science and Innovations
George Daley on Korean stem cells' true source
Research led by George Daley, MD, PhD, and Kitai Kim, PhD, sheds new light on the now-discredited Korean embryonic stem cell lines (Hwang Woo-Suk et al.), setting the historical record straight and also establishing a much-needed set of standards for characterizing human embryonic stem cells. In this interview, conducted on July 20, 2007, Dr. Daley discusses the Korean research and how the Korean stem cells were actually created through parthenogenesis, a process that creates an embryo from an unfertilized egg.
[Go to the press release...]
What is parthenogenesis, and how does it figure into the origin of the discredited Korean embryonic stem cell lines? [3.6 MB]
How did you figure out the Korean cell line's origins? [4.0 MB]
And what genetic signature did you find in parthenogenetic stem cells? [5.2 MB]
Have parthenogenetic embryonic stem cells ever been made before? [0.8 MB]
Why are parthenogenetic embryonic stem cells important? [2.6 MB]
How would this bank of parthenogenetic embryonic stem cells work? [3.9 MB]
Does parthenogenesis mean we no longer need to work with IVF embryos to get stem cells? [1.9 MB]
Does generating embryonic stem cells through parthenogenesis avoid the ethical concern about destroying embryos? [2.0 MB]
[Note: The free QuickTime player is required to view these video clips. If you can hear the audio but see no picture, you may need to install a more recent version of QuickTime.]
1. What is parthenogenesis, and how does it figure into the origin of the discredited Korean embryonic stem cell lines?
Embryonic stem cells can be made by a number of different techniques. Most embryonic stem cells are derived from in vitro fertilization embryos. So [with] couples who are undergoing assisted reproduction, the egg and sperm [are] brought together in a Petri dish, and then from the embryos you derive embryonic stem cells.
The second strategy is to use nuclear transfer. This is where you would take an individual's cell, and you would take its DNA and put it into an egg whose own DNA has been removed. That will then create an embryonic stem cell that's a perfect genetic match for the donor.
The third way we make stem cells is by this process of parthenogenesis. This is a fascinating process where the egg is activated without fertilization, so there's no contribution of sperm. It's tricked, if you will, into developing into an embryo. It can never develop into a full organism, but it can make embryonic stem cells.
We know now with great certainty that the first supposed nuclear transfer line from the Koreans is indeed a parthenogenically derived line. It came entirely from the cells of the woman who donated the eggs. Now in that case, one reason that's interesting, is in that original experiment, they took the donor cells, the skin cells, of the same woman who was also donating the eggs. So in that process, it actually was quite difficult to know whether the resulting cell line had actually come from the skin cell or from the egg itself. And that's what created much of the confusion because they should be essentially genetically identical and indeed they are, but what differs is the preservation of certain regions of the chromosomes and the way they recombine. That's a signature that unequivocally can distinguish whether the cell line was derived by nuclear transfer or cloning, or by this process of parthenogenesis.
2. How did you figure out the Korean cell line's origins?
We discovered fortuitously that we could quite easily distinguish an embryonic stem cell that had come from a naturally fertilized embryo or a somatic cell nuclear transfer embryo from those that have been generated by parthenogenesis, without the contribution of fertilization. That's critical because in attempting to make cell lines by nuclear transfer, one can actually, by mistake, isolate parthenogenetic lines.
In a typical somatic cell, you'll have both a copy of the mother's gene and the father's gene. Whereas in a parthenote, you'll have two copies of the mother's and zero copies of the father's.
The predominant misconception in the scientific community was that a cell line that was derived by parthenogenesis should be what's called homozygous. That is, it should have taken the copy of the mother's DNA and duplicated it entirely, so that all the genes all along were identical. And yet we know from our work that that's actually a misconception, that and in fact, the way most parthenogenesis is done leaves the cell line with actually a majority of mixed up genes from the mother and father, what we call heterozygosity.
When you're trying to determine whether two biologic samples are related, you look for these so-called single nucleotide polymorphisms, or these subtle genetic changes, that distinguish the mother's genes and the father's genes. What we've done in the past is look at a relatively small set of those markers -- maybe three or four, and when you're only looking at three or four markers, you don't have as much resolution. You don't have the ability to say with great confidence that two samples are related or unrelated.
What has come through the Human Genome Project is now the availability of thousands and thousands of these markers. So you can actually analyze across the entire genome -- in our case, 50,000 [correction: 500,000] different markers, which gives you an enormous amount of genetic resolution.
3. And what genetic signature did you find in parthenogenetic stem cells?
In order to understand how the genetic signature can be different, one has to think about how the egg matures and how the chromosomes within the egg actually get rearranged. So I'm going to use my fingers as chromosomes. Typically, we inherit one chromosome from our mother and one chromosome from our father. And then when a cell goes through division, those chromosomes duplicate. When they duplicate, they're attached by a little spot, a little stalk, that's called a centromere. When a cell goes through mitosis, what happens is the centromeres split, and the chromosomes segregate so that that each daughter cell gets one of the mother's and one of the father's chromosomes. In meiosis, where you make a germ cell -- or gamete, a sperm or an egg -- there's a difference. The mother's and the father's chromosomes duplicate, but then they pair and the homologues, or similar, chromosomes pair, and during that pairing you get genetic recombination; rearrangement. Now, if I could paint two of my knuckles black, you'd see that I could transfer a black knuckle over here. And what happens in meiosis is you get two segregations of the chromosomes. In the first segregation, you actually have the original mother and father chromosomes separated. What's left behind is a duplicated copy of either the mother or the father's single chromosome. But now, remember, one of these might have a black piece.
So it's actually not the exact same chromosome as the mother. One is the same, and then the other has a piece of the father's DNA. And you can see that genetically. So when you make a parthenogenetic line, what you do is you block the segregation of these chromosomes, so that the two original chromosomes are kept in the cell. Now parts of the chromosome-especially the part around that little knob, the centromere, they're less likely to have recombined. So they would tend to have the same genes, whereas as you get farther away on the chromosome, it's much more likely that they'd recombine. That signature, then, becomes very, very obvious. As you follow the markers along the chromosomes, the one near the center are the same, but as you get further and further from the center, they're more likely to be different. And that's unique to the process of parthenogenesis. There's no other way that chromosomes segregate in biology that would have that same genetic signature. That's what allows us to know whether or not the cell line has been derived solely from the egg's chromosomes, or actually could have been derived by the process of nuclear transfer or cloning.
4. Have parthenogenetic embryonic stem cells ever been made before?
Embryonic stem cells have been made from mouse embryos for decades, and for the last few years, we've actually been able to make embryonic stem cells from monkey embryos that have been generated parthenogenetically. But it's been more challenging to get that to happen in humans. In fact, the Hwang line teaches us that you can indeed make human embryonic parthenogenetic stem cells. That's highly valuable.
5. Why are parthenogenetic embryonic stem cells important?
One of our goals in this in this entire area of work is to identify how to make embryonic stem cells that have the tissue type of a particular patient. We'd like to make rejection-proof embryonic stem cells. One strategy is nuclear transfer, which is this method of bringing the patient's cells into an egg whose own DNA has been removed, creating essentially a rejection-proof customized cell. However, that process has not been reduced to practice in humans. It's very inefficient, even in the various mouse models, where we do it quite readily. Parthenogenesis, on the other hand, is much more efficient. We've explored whether we can generate healthy and functional embryonic stem cells using parthenogenesis. That has, in fact, been done by many groups for decades. But the twist we've added is to use genetic selection to identify those parthenogenetic cells that would be a perfect tissue match at the tissue rejection antigens to the egg donor. You could also select the cells in such a way that they might match the egg donor's siblings, or a very significant percentage of the population. For the first time, we have a prospect for making an embryonic stem cell bank that would have a reasonable probability of matching a significant percentage of possible transplant recipients in the community. That's really what's driving this work.
6. How would this bank of parthenogenetic embryonic stem cells work?
By doing this genetic analysis, you can determine directly the genes at the tissue-matching loci. So, when we transplant organs, typically the immune system focuses on a very small number of genes, and they use those genes to either see the tissue as self or to reject it as foreign. So when we do bone marrow transplants, for instance, we are matching those genes. Typically, we worry about three genes from the mother, and three genes from the father. For a bone marrow match, you need to have six out of six.
The process of parthenogenesis, if we remember, can duplicate DNA, a chromosomal piece. And so if they duplicated the place where the tissue-matching genes are located, now instead of having six different genes, you actually have three in duplicate. So it's much easier to match people in the population if you're only worried about making three identical rather than six identical. So for the first time, this raises the prospect that you could, with a relatively small number of master cell lines, selected for the fact that they're only carrying the three tissue matching genes, that uou could make a bank of those and have all the major common sets of three -- in fact, that's been modeled in a paper in Lancet in the last couple of years -- that you would with as few as 10 of these cell lines be able to have a tissue match productively for up to 80% of the population.
I think the implications for imagining a future where we would create master banks of embryonic stem cells and have an off-the-shelf type of therapy -- off the shelf cell products -- rather than the highly cumbersome, highly labor-intensive, highly expensive, individualized patient-specific therapy, I think has been an advanced by the idea that you can couple parthonogenics and genetic screening to identify those cell lines that are going to be most helpful.
7. Does parthenogenesis mean we no longer need to work with IVF embryos to get stem cells?
The fact that we can generate embryonic stem cells from eggs alone does not obviate the need for generating embryonic stem cells from leftover embryos from fertility clinics or by the process of nuclear transfer. Each of the strategies has its own applications, and there are certain types of research and certain fundamental questions -- and major areas of therapy -- that can only be addressed by these other types of embryonic stem cells.
For instance, if we want to treat someone of a particular ethnicity, where we won't find an easy match in a parthenote bank, or we want to treat an individual with a specific genetic disease, in that case having your own cells is always going to be better.
So creating these cell banks, using parthenote tissue, is a compromise. It's not as good. If we could do it in a facile way, in an efficient way, and in an inexpensive way, and be able to treat individuals easily, we would always go for individualized therapy. But that may not be practical in the long run.
8. Does generating embryonic stem cells through parthenogenesis avoid the ethical concern about destroying embryos?
There has been a lot of debate about whether creating cell lines from eggs alone that don't involve the destruction of a fertilized embryo -- an embryo that would have full human potential -- that this may skirt some of the ethical concerns. I think it will for some people. The fact is that we're still creating a structure that looks like an early human embryo. I think it's interesting because that early human embryo does not have any full developmental potential. We know that from looking at this in lots of different mammals. We know it in monkeys and we know it in mice. So this is a cluster of cells with no human potential, and yet has enormous potential to generate cell lines.
I happen to think that working with these cells should be very ethically uncontroversial. But for those who are absolutist about the nature of the embryo -- who see it as a cluster of human cells -- then the controversy may persist.