THURSDAY, July 24, 2003
Session 3: Stem Cell Research: Recent Scientific and Clinical
John Gearhart, Ph.D.,
Professor, Institute for Cellular Engineering,
Johns Hopkins University
Rudolf Jaenisch, M.D.,
Member, Whitehead Institute;
Professor of Biology,
Massachusetts Institute of Technology
David Prentice, Ph.D.,
Professor, Life Sciences,
Indiana State University;
Adjunct Professor, Medical and Molecular Genetics,
Indiana University School of Medicine
CHAIRMAN KASS: Could we come to order, please? The third
session of this meeting is on "Stem Cell Research: Recent Scientific
and Clinical Developments." The most challenging aspect of
trying to fulfill our charge of monitoring stem cell research is
the monitoring of the scientific research itself. There is so much
of it, it's very diverse with many sources and types of stem
cells, and many types of research. Things are changing amazingly
rapidly, and we are trying to monitor a moving target. The material
is highly technical and hard, even for scientists outside of the
field, let alone a layman, to evaluate carefully.
"The Isolation of Human Embryonic Stem Cells," and,
"Human Embryonic Germinal Cells," by James Thomson and
John Gearhart respectively was reported only five years ago. The
obvious promise of these cells both for gaining knowledge of development,
both normal and abnormal, and eventually for regenerative therapies
with hundreds of thousands of patients with degenerative diseases
or injuries has produced great excitement, much fine work, and to
be frank, more than a little hype.
At the same time, reports of previously unknown and unexpected
multipotent or stem cells in various tissues of children and adults
has produced enormous interest also in so-called adult stem cells,
with similar promise, and to be frank, also more than a little hype.
By almost all accounts we stand today in the infancy, not to say
embryonic, stage of these researches, and it is surely too early
to answer the questions that the layman wants answered. How soon,
for which of our diseases, from what sort of cells, at what cost,
and at what risk will the cure be available?
Yet it is not too early to learn where we are in fact in this
rapidly growing field, to try to separate fact from fiction, true
promise from hype. And to do this we have gone to the experts,
commissioning review essays, essays that would review for us the
published literature over the past two years since the August 2001
decision, covering work in five areas according to the origin of
the cells. "Human Embryonic Stem Cells," a paper by Tenneille
Ludwig and James Thomson, "Human Embryonic Germinal Cells,"
a paper from John Gearhart, a paper on cloned embryonic stem cells,
a paper by Rudolph Jaenisch, a review of the adult stem cell research,
a paper from David Prentice, and an update on the work with her
own multipotent adult progenitor cells from Catherine Verfaillie.
The papers have been sent out with the briefing books, and I assume
that they've been read. We are thinking about another paper
on mesenchymal stem cells.
We're very fortunate and grateful to have with us this afternoon
three of the authors, Dr. John Gearhart, Dr. Rudolph Jaenisch, and
Dr. David Prentice, who will, in brief presentations, highlight
their own papers, after which we will have questions and discussion.
I would remind Council members that the purpose of this session
is wholly scientific. We want to learn as much as we can about
the current and projected state of this research. This isn't
the time for ongoing ethical arguments about the moral status of
the embryo, the research imperative, or equal access for the current
funding policy, important though these issues are. And I will simply
say I'll use the authority of the Chair to see to it that we
try to stay on the topic.
Dr. Gearhart has kindly agreed, by the way, to field questions
pertinent to the review essay submitted by Drs. Ludwig and Thomson.
Gentlemen, welcome to you all. Thank you for your papers, your
presence, and in advance for your presentation and participation.
And we will proceed in alphabetical order with John Gearhart first.
DR. GEARHART: Well, I can't say I'm delighted
to be back, but I'm happy to be back. And to not really dwell
on the research that's going on specifically in our laboratory,
but to give you a bit of a general overview of where the work is
in this field, and I think there will be ample opportunity from
some of the things that I say that you can ask questions about it.
I think the first question, and judging from some of the morning
topics, that I would like to address is the one before you now,
which is, why study these cells? And what we have heard most frequently
are the issues, well, this is going to be the most versatile source
of cells for promised cell-based interventions for various therapies.
But I came at this at a different angle. I am an embryologist
by training. I've been a student of human embryology for 28
years. And I'm extremely interested in knowing how we are put
together. What are the events, what are the mechanisms that lead
to the formation of the human being?
And over time, as you can imagine, we've studied fruit flies,
we've studied mice, we've studied rhesus monkeys. And I
think one of the major, and perhaps the most important bit of information
that we're going to get out of embryonic stem cells are going
to be the issue of how we are formed. And this is going to involve
cell biology, genetics, on and on and on. But this is what is going
to enable us. And I think with this information, which at the moment
is under the heading of Fundamental Science, or Fundamental Discovery
in Basic Science, this, I think, is going to be the lasting contribution
of studies from embryonic stem cells.
They're going to have applications, obviously, with that information,
and the sources of cells. And we're going to apply those to
birth defects. We're going to apply it to injuries and disease,
because many of the processes are similar. And we're going
to be able to draw on that information in trying to correct diseases
True, I think out of this we're going to get sources of cells
which, in the short term, are going to be used directly for the
cell-based interventions. And I think through the use of these
cells, we're going to discover other important things about
these cells, and some in our research we've learned, which we
can apply in therapies.
But I want to emphasize one point, and this is my belief, that
this period of time that we're going through in studying embryonic
stem cells is going to be transient. And I want to emphasize that
I believe it's the information we're going to get out of
these cells that are going to be used in the longer term to restore
function, to regenerate tissues within the bodies of patients, without
the use of these cells. And that to me, as I look downstream, is
going to be the most important outcome of this research.
Now I realize that this quote is taken out of context for these
essays that were written, and what was being questioned here. But
I want to make some comments which are important when we consider
the use of human embryonic stem cells directly. And that is, we
are finding enormous differences between mouse embryonic stem cells
and human embryonic stem cells, not only in practical issues within
the laboratory from the standpoint of cell cycle times, et cetera,
but also, as we begin to differentiate these cells in a dish, we
find some very subtle and some very not so subtle differences between
the mouse cells and the human cells to get us to the differentiated
functional tissue that we need.
So I do believe it's time that we work directly on the human
embryonic stem cells if we want to gain this information on embryogenesis,
and if we want to move forward with developing cell-based interventions.
So, the sources that we're going to talk about in general
deal with this very early stage here derived from the IVF-donated
embryos, and from tissues collected within the early stage of fetal
development, the germ cells. These cells have many things in common,
and we've published on this. You've seen this, probably.
And so we look at them as a group.
There's a third member of this group, which is embryonal carcinoma
cell, depicted here on the right, which has very similar properties
to these two other cell types. In fact, it's the only cell
type from which any cell has been derived that's being used
in a pre-clinical trial. And this is on stroke patients at the
University of Pittsburgh.
I want to spend time now, or my brief time, on three slides to
give you an indication of where the human work is on this class
of cells. What are the questions that are being asked, where are
we on some of this stuff, et cetera?
So here now are issues around human embryonic stem cells, or embryonic
germ cells. One of the first questions that's being asked about
the stem cell itself is what is meant by "stemness"?
And obviously this, at the molecular level, gets us into genomics,
proteomics. And so there is a major effort around the world at
trying to determine the genetic basis of a stem cell. And the human
cells are being used extensively for this purpose. How many genes
are involved, what genes are involved, what signaling pathways,
on and on and on. So stemness is an important issue within this
Now, it's a comparative thing. We can do it also with mouse
cells. We can do it with adult cells as well, trying to find commonality.
What is it that makes a cell a stem cell? That information will
be vital, I believe, in the future, in converting virtually any
cell to a stem cell, if we want to go that route.
The second major issue currently are characterizations of existing
lines. As you may know, there's no standardization within the
field as to what constitutes a human embryonic stem cell. We talk
about developmental potential. We talk about molecular markers,
biochemical/antigenic markers, this is now being taken up by an
international committee under the auspices of the MRC in London
to try to get an agreement on these issues from stem cells around
the world, human embryonic stem cells around the world.
There's also the issue now of whether or not we should be
deriving new stem cell lines. And in many countries, this is indeed
happening. There are a number of new cell lines coming online.
And what is the importance of this? Now, many of you have heard
at least from what the Europeans refer to as the presidential cell
lines, those that are eligible for funding in this country. Should
we have cells that are only grown on human feeder layers? Most
of the cell lines that have been derived are on mouse feeder layers,
and there is a concern about infectious agents passing into the
I think there's a misconception out there that these lines
will be ineligible for any kind of therapeutic use. I think the
FDA has made it very clear that they're certainly willing to
consider those grown on mouse feeder layers if they meet certain
criteria to also be eligible for use in the clinic.
But there are a number of lines now that have been derived on
human feeder layers, and there are lines being derived that are
feeder layer independent, which means it makes it much easier to
work with in the laboratory. You don't have to worry about
these other cells that are present in your cultures.
And there's an attempt now being made to derive these lines
in defined media, which means there are no other animal products,
no serum, that you can control very carefully the differentiation
of these cells by adding specific growth factors in a sequential
The most important element to this, though, is this top line here.
Many of you may not realize it, but when we say that routinely three
or four different mouse embryonic stem cell lines have been used
for the last 20-some years, what's the problem here, why can't
we use just a few of the human. Those lines have been selected
from a hundred and some cell lines that were made as to being the
best cell lines to use for those purposes. And so I think we're
still in this phase of trying to determine what are really the good
cell lines, human embryonic stem cell lines that are available.
And thus the issue of generating more.
For the use of these lines, emphasis in the lab now is in the
following area. We're trying to work up high-efficiency differentiation
protocols that result in homogenous cell populations of functional
cell types. That is, can we in the laboratory come up with protocols,
growth conditions, that are going to get us 10 million dopaminergic
neurons, and only dopaminergic neurons? Most of the protocols we
have now result in a mixture of cell types within a dish, and we
go in and select out the ones we want.
So there's concern about the heterogeneity of the cells, and
the fact that we can't at this point have a high efficiency
in getting the cell types we need. We manipulate the growth environments,
and we manipulate the cells genetically to accomplish this goal.
Once we differentiate these cells, and we've had some success
in certain pathways which we can talk about, we are very concerned
about what we refer to as the authenticity of the derived cells.
That is, is the dopaminergic neuron you've formed, the insulin-producing
cell you've formed, the cholinergic neuron, the cardiac muscle,
are these actually functional cells that are the equivalent of those
that you normally find in the adult?
So a great deal of effort, both in the dish, in vitro,
is being made in different parameters to measure whether or not
these are authentic cells. There are many protocols, as you can
imagine, that lead to cells that look like something, but they're
not functional. And this is a major concern. The other side of
this is, the golden test is can you transplant these cells into
an animal model of some kind and see that they assume the function
that you're hoping that they would.
But again, these are human cells going into rodents, primarily.
Do we have every reason to believe that they're going to function
appropriately? We don't know that yet. We're also putting
cells into non-human primates to get that kind of a measure as well.
We are in the laboratory contrasting and comparing among various
sources of stem cells. In our laboratory, we work on human ES,
human EG, umbilical cord blood, et cetera. And the importance of
this is that you are able to compare and contrast in the same paradigms
the sources of these cells to see which ones are going to be appropriate
for which tasks. And I think this is extremely important.
We're also very concerned, as you know, if these cells are
going to be continuously grown in the laboratory, there's a
finite possibility that you're going to get genetic mutations.
We're interested in the frequencies of this, and the types of
the mutations that occur in these cells. It's a safety issue.
But with our genomic and proteomic studies, we can get a good handle
on this at this point in time.
When we get into the grafting issues, which is obviously a desirable
endpoint, both in the basic science side, to say you have a functional
cell, and then transitioning into any use of these cells in cell-based
therapy, we have a number of issues here.
The first, and I think you have to realize this. There is virtually
no appropriate animal model for any human condition. That's
the fact. We approximate it. We do the best we can. But under
those limitations, we do the best we can. And you have to be aware
of this. There's a concern about what animal models you're
using for what test.
The issue of what stage of cell differentiation in any of these
cells do you use for a specific graft? We are finding that many
things within the central nervous system, you want to use something
that is fairly undifferentiated so that the cells can interestingly
migrate to where they're supposed to go, set up the appropriate
connectivity. And so if you are working with a motor neuron, you'd
like to use a motor neuron precursor. If you put in a fully differentiated
motor neuron, nothing works. I mean, when you think about the nervous
system, it's not surprising.
With insulin-producing cells, we find that you want the terminal
cell, at least in the experiments that we are doing. But for each
of the cell types you're working with, we've got to determine
this, as well as how do you want to deliver it to those animal models.
The last two are extremely important. What are the fates of these
cells that you've grown in a dish, selected for in a dish, once
you graft it into an animal paradigm? We know these cells like
to migrate. Do they differentiate appropriately? Do they form
tumors? I remember a few years ago when I was all puffed up that
our cells were working so wonderfully, and I went to the FDA for
this meeting I thought was going to be one-on-one with members of
that group. And it turned out that there were 40 of them and one
of me. All their questions were about safety issues. We now spend
as much time on these safety issues as we do at looking at differentiations
of these cells in a dish.
Why? Well, what have we found? Indeed, in the early mouse work
with embryonic stem cells the rates of tumors were extremely high.
Formation of teratocarcinomas, mixed germ cell tumors. We have
at this point, though, grafted well over 3,000 animals with our
human cells, and we haven't seen a tumor yet. And maybe we're
just lucky, maybe there's a difference between mouse and human
cells. We don't know that.
But these experiments are extremely expensive because you've
got to serial section through the whole central nervous system if
you're putting cells in there. You're putting them in IP,
et cetera. It's a lot of work. The FDA, ideally, would like
to know if you're putting 300,000 cells into an animal, they
want to know where all 300,000 cells went. This has gotten us into
extremely expensive experiments in labeling cells, developing labels
for cells that you could follow for a year or more later. You could
go back and find those cells. So we've employed radiologists,
and chemists, et cetera, to help us with these kinds of things.
So it is a major issue. And in all of our experiments, we look
The last issue, on immune response graft rejection, is obviously
to some extent the Achilles' heel of this work. If we can't
provide tissue that's going to be accepted in a graft, then
all this work has gone for naught. So there are issues of active
experiments now going on in tolerance, in genetic alterations of
cells, the consideration of cell banks, how many different stem
cell lines would you need to cover certain populations of people,
and something you'll hear about probably from Rudy on nuclear
and cell reprogramming, so-called therapeutic cloning. Is this
one of the avenues that we can approach? So all of these are very
active research elements going on with respect to the human embryonic
germ cells and stem cells. We can't divorce them.
Now, in this cartoon, what I've depicted here is one of the
more difficult scientific aspects of the research. And that is
that you have in a dish—now this is a bit of a repeat, but it's
important to mention—you have in a dish these cells that are capable
of forming any cell type that's present in the body. The major
problem here is how do you get them only to form hepatocytes, dopaminergic
neurons, heart muscle, blood cells, pancreatic islet cells, solely.
How do you do that? This is where the real issues are within the
And we try to recapitulate, as I showed you this before—oh,
and finally, the issue of—I put this little thing down here to
remind me. Some of you have seen this report earlier this year
from Hans Schuler's lab at Penn, that he's been able to
find oocytes within embryonic stem cells derived from the mouse.
Now, he did not—now keep this in mind, he did not purposefully
differentiate these cells into oocytes. He came up with a neat
little trick of identifying them once they were present in the plate.
Okay, so it's not that he's worked up a condition to differentiate
them only to oocytes, it's that they're there. You can
use that technology to pull the cells out, and then further try
to mature them, see if they are actually functional oocytes, et
So we rely in the lab on trying to recapitulate what has occurred
to some extent in the embryo by providing growth factors that these
cells normally see to get them to enhance, or try to direct their
differentiation into the product that we want.
We use selection techniques in the laboratory after getting these
mixtures of cell types by different avenues in the dish. We then
come back using genetically selected markers, different growth media,
cell sorting, which is a fancy way of sending cells through a beam
of light, and if they have a certain antigen on the surface, you
tag it in a way that will put them into different pots. And this
works fairly effectively for a number of cell types. But it's
highly inefficient, and we've got to improve this technology.
So what have we done? What have we been able to do? We've
been able to identify many, many different cell types in a dish.
We've been able to expand many different cell types, whether
they're insulin-producing cells, dopaminergic neurons, heart
muscle cells, smooth muscle cells, hepatocytes, we're able to
do that. Human. And we've been able, in many cases, to show
that they are functional within the parameters that we like. And
in a few cases we've studied extensively we've been able
to graft them into animal models to show that they will work, and
can ameliorate a condition in the animal, whether it's, to some
degree, whether it's a diabetic animal, or one with motor neuron
loss, mouse models of Parkinson's disease, we can do that at
To reiterate, we are concerned about safety issues. They go hand
in hand with all that we do in the laboratory. It adds, again,
expense and time, but we want to make sure we get it right.
Here's another issue of safety. We find that in many of these
cell lines, if you're not careful—now human embryonic stem
cells are much more difficult to grow in the laboratory than mouse,
and if you're not careful, this can overwhelm your cultures
very quickly. This is an example of a tetrasomy 12p, and also of
a trisomy 17, which appear to be fairly common coming out of the
human cell lines for whatever reason. We don't know. But
you have to constantly monitor these cell lines to make sure that
they are remaining normal as far as the karyotype is concerned.
Not an unusual circumstance.
Rudy, I'm sure, will deal with this. But we got highly involved
with the issue of somatic cell nuclear transfer, mainly coming in
from the desire to have matched tissue for a patient. But I want
to point out two other issues here that I think, in my way of thinking,
are perhaps more important. Rudy will probably disagree.
Number two on this list, to increase the diversity of stem cell
lines. Number three, to facilitate the study of inherited genetic
diseases and somatic mutations. Now, there's been a lot of
talk about, with the justice issues, of diversity of cell lines.
We now know how many cell lines we need to cover the population
of the United States. How are we going to get these lines? We're
not going to get them, if we wanted them, by going out and matching
up people to get embryos, okay, or even assaying embryos in banks.
The easiest way to do it is you identify a person that has the haplotypes
of class I and class II genes, and take a nucleus out of their cell.
And generate stem cells through somatic cell nuclear transfer.
No question about it, easiest fast way of doing it.
The issue of the studying of inherited genetic mutations and somatic
mutations. It means that we can take cells from a patient with
a disease, some of them not expressing very much from the standpoint
of the pathogenesis of the disease, to others that are full-blown,
to try to figure out why some patients have one form of the disease,
another has the mild form of the disease. We can take this directly,
and generate stem cells for those studies. Somatic mutations, breast
cancer mutations. We can generate cell lines—it's the only
way we can do it—through somatic cell nuclear transfer.
And then the issue of being able to determine mechanisms of nuclear
reprogramming. We would hope that they would approach normalcy
from the standpoint that we may be able to use that information
to convert somatic cells into stem cells at some point downstream.
This is why I think somatic cell nuclear transfer into a human are
reasons that are quite important.
Well, I think with the issue of the human embryonic stem cells,
I think they have uniqueness in certain areas, and I think that
what we're going to learn out of these studies is going to be
extremely important in the future, not only for deriving cell-based
therapies, but in getting information that we're going to be
able to begin to instruct our own genes.
And this is something which hasn't been paid a lot of attention,
but I think even within this Council, it may be a good idea, that
we're on the verge, I believe, of being able to instruct our
own cells. And I think the ramifications of this are enormous.
I mean, talk about enhancement. One could go on the Internet and
get a kit for enhancing any part of your body to a certain degree.
And this is all going to be done, I think, from information that
we get out of learning about cell differentiation, how these cells
function. So I think we've got to keep our eye on this for
the future as well.
Just in closing, the greatest impetus that I can think of for
making advancement in this field is funding through the National
Institutes of Health. This is the only information that I could
get my hands on. It's available. It's for the fiscal year
2002. You see how much money is being spent on stem cell research
in toto, how much on human adult stem cell research, how much on
animal adult. And up to this point, $11 million. On the human
embryonic stem cell research, I think this year it's going to
be more, and we would hope in the future, as the bottleneck for
cell lines and more applications come in to the NIH, that these
numbers will become robust.
And I think this will be the key to the dreams that we have, anyway,
of learning more about how humans are built, and cell differentiation
of human cells on their way to therapy. So I hope I've given
you enough to think about. Thanks.
CHAIRMAN KASS: Thank you very much. Why don't we
simply proceed in order. Dr. Jaenisch.
DR. JAENISCH: Well, thank you very much for inviting me
here. And I will right away get to the cont roversy. I'm not
going to agree with much you said.
So I want to try to develop two arguments. One is reproductive
cloning faces principle biological problems that may not be solvable
for the foreseeable future. And the other one is therapeutic cloning
poses no principle biological obstacles, only technical problems.
And this, I think, in my opinion has implications for the status
of the fertilized and the cloned embryo.
So you need one piece of data. When we do nuclear transfer, then
it's clear that the donor cell, the donor matters. So if we
use a somatic donor nucleus, cumulus cells, fibroblasts, Sertoli
cells, the result to get cloned adults is really low, a few percent.
If you use B or T cells, or neurons, it is very low. Only with
tricks we were able to get animals. However, when one uses embryonic
stem cells as donors, it's an order of magnitude more efficient.
So the conclusion is an embryonic cell is easier to reprogram to
support life than a somatic one. I will come back to this later
when I make an argument.
So, let me first come to cloned animals. If you have an animal
born, if you looked at this carefully, you find that approximately
four to five percent of all genes, as done in percentile, are not
correctly expressed. Still, the animal develops to birth. If you
look at imprinted genes, it's even worse. Thirty to fifty percent
are not correctly expressed, regardless what the donor cell nucleus
is. So from this we would argue that it's amazing that these
defective embryos can develop to birth and beyond. But you pay
for that. You pay with abnormalities.
So my argument would be, and I will develop this more, that there
may be no normal clones. Of course, we have the problem to define
what we mean with "normal". But let me then say that
faulty reprogramming seems to me is a biological barrier that may
preclude the generation of normal cloned individuals, at least for
the foreseeable future.
Let me come to therapeutic cloning. So, as we heard, cloning
would involve taking from a patient a somatic nucleus and derive
an isogenic embryonic stem cell by nuclear transfer. This could
be used, then, to correct the genetic defect if it was present.
One has to differentiate these cells in vitro, and then transplant
it back into the individual and see whether it could improve the
So we have made such an experiment. I just want to briefly outline
the steps we went through. As a patient we used a mouse, which
was totally immune-incompetent. Due to the Rag2 defect, it did
not have any B and T cells. So that was our patient.
From this, skin cells were removed from this animal. They were
transferred. A blastocyst was derived. And from this blastocyst,
an embryonic stem cell, which of course is mutant, as the donor.
And then we used just simple homologous recombination to repair
one or two alleles to make a functional Rag2 gene, and then differentiate
these cells which had problems into hematopoietic stem cells, put
them back into the animal, and the result was that these cells could
colonize the embryo and repair, in part, the defect.
So some have argued that this experiment really didn't work.
It really only showed that when you have—I think Dr. Prentice
argued that it only showed that one has to use a cloned newborn
as a source. And this, of course, is not therapeutic cloning.
I think that is a misrepresentation of those data.
I would argue that from this experiment we know that somatic nuclear
transfer and therapeutic cloning will work because it has worked
in the mouse. So only technical, not principle, barriers exist
to adapt it to human use, although these technical ones may be daunting.
Technical, in contrast to principle ones, which we see in reproductive
So, if I look at the two ways we can think about cell therapy,
then we have the nuclear cloning approach as I just outlined, which
in the mouse we know works, at least for hematopoietic cells, and
I believe will work for other cells. The alternative is somatic
adult stem cells. We'll hear later from this.
And I believe here there are many, many question marks. It is
a very young field, very interesting, but we really don't understand
how to really handle these stem cells, how to propagate those.
And there's really, with the exception of bone marrow stem cells,
not really any proof of therapeutic potential at this point. So
I think we need much more to learn here, but this we know will work.
So I want to come to really two key questions. One is can reproductive
cloning be made as safe as in vitro fertilization? And the
other one: Does faulty reprogramming after nuclear transfer, does
it pose a problem for the therapeutic application of this technology?
I think these are the key questions in my mind.
So, can reproductive cloning be made safe? I believe in addition
to the technical problems, which are solvable, there are serious
biological barriers. And the main one is this one. The two parental
genomes in all of our cells are differentially modified depending
on where the genome came through the egg or through the sperm.
They're epigenetically distinct in the adult. If you want to
recreate that situation, you would have to physically separate the
two genomes and treat them independently in an oocyte- or sperm-specific
So let me point that out in this very complicated-appearing diagram,
but it's very simple. So let me just go through briefly what's
happening during normal fertilization. In normal fertilization,
the egg genome and the oocyte genome are combined. And they are
differentially modified during gametogenesis. And this is what
these lollipops here show. So they're different in the zygote.
But now something very curious happens. Within hours of fertilization,
the sperm genome is stripped of all methylation. The oocyte is
resistant to this not-well-understood demethylation activity because
the oocyte genome has been, of course, together with this activity
So then cleavage proceeds. And the point I'm making now is
that the cells of the genomes in the adult are different, different
because of their history. It's not only for imprinted genes,
but also for non-imprinted genes. Now what happens in cloning?
In cloning, one removes, of course, the egg nucleus and replaces
it now with one of the somatic nuclei.
Now, both of these genomes are now exposed to the egg cytoplasm.
So both become really—they're in identical chromatin state.
So they'll be both modified, and the difference will be equalized.
So the difference you see in a normal adult cell will be equalized,
tends to be equalized in cloning. I think that is a problem among
those which we call epigenetic problems.
So in order to solve that, I think one would have to separate
the two genomes physically and treat one in an egg-specific appropriate
manner, and the other one in a sperm-specific manner. I think if
we could do that, then I think we would have a way to solve that
So the phenotypes of clones, I would argue, is a continuum without
defined stages. So we know the most important stages are implantation
and birth, and indeed, most clones are lost at implantation, and
then at birth again many are lost. You end up with very few which
develop to late age. And we know from most experiments, even animals
which are one year of age, 80 percent of those die very early with
major problems. These experiments could only be done so far in
the mouse because the mouse is the only organism we have old cloned
So I think the problem here is, yes, you might get—because our
criteria are not very good here—you might get quasi-normal or
maybe a normal individual with a certain frequency. It's very
difficult in animals to test, really, because our tests are limited.
But we cannot predict ever at any of these stages who will be among
those outliers, and who will not be. There's no way to select
good healthy from non-healthy clones. I think that's very important
to emphasize. There's no way we can think of to do that. So
there's no predictability whatsoever. So from my point of view,
this doesn't matter, really, whether one or two percent of potentially
normal animals which survive for a long time. You can't predict
who this will be.
So, therapeutic cloning. So, as I said, cloned animals are abnormal
due to faulty reprogramming. Why is normalcy of differentiated
cells derived from ES cells not affected by this problem? There
are two reasons. One is that we don't involve a generation
of a fetus. And the other one is that the embryonic stem cells
lose what I'm going to call epigenetic memory of the nucleus
they came from. So let me develop these two ideas.
So first, it's simple. In contrast to reproductive cloning,
there's no embryogenesis required to derive functional cells
in vitro. So it's totally irrelevant, for example, whether
imprinted genes are reprogrammed or not. Imprinted genes have a
function only during fetal development, not in the adult. So it
doesn't matter. There is one or two exceptions, and those would
have to look at.
Very importantly is the cells themselves select themselves in
vitro for the cell type you are selecting for. There's
no selection possible in vivo after implantation in a cloned
animal. And then we know that cloned ES cells form normal chimeras,
as do any other ES cells. So biologically there is no difference.
Let me come to the memory question on this slide here. So we
know we can make blastocysts from a zygote from a fertilized egg,
or by nuclear transfer from these various donor cells. Now I would
argue the blastocyst remembers exactly where it came from. And
we know that because we implant this blastocyst, derived from a
zygote, into the uterus, it will with a high efficiency develop
to birth, and will make a normal animal.
If you derive—take a blastocyst derived from an ES cell, and
implant it, it will develop with a high efficiency to the newborn
stage, but it will make an abnormal animal. If it came from a cumulus
cell, it would be very low efficiency—I showed you this on the
second slide—low efficiency to the newborn stage. It will be
abnormal. If the nucleus comes from B-, T-cells, or neurons, it
will be very low. And it will be, of course, abnormal, if you get
So these blastocysts know exactly where they came from. And we
know this, actually, when we look at gene expression patterns in
clones derived from ES cells, or from human cells, they're different.
So they remember at birth where they came from. That's why
cloning doesn't work. That's why these cloned ones develop
abnormally, too, in the great majority.
So what about ES cells? In ES cells, it's a very different
thing. I would argue you erase the epigenetic memory of your donor
nucleus. And the argument is the following. We know that if you
take such a blastocyst, the ES cells are derived from the inner
cell mast cells. These inner cell mast cells express certain set
of genes, one of those called the Oct-4 gene, one of the key genes,
but another 70. You put this blastocyst in culture. All these
genes, all these cells array—silence, Oct-4. This has been published.
And they don't divide. They sit there.
But over the next days or week or so, some of these cells, one
or the other, begins to re-express Oct-4 and another set of 70 genes.
And these are the cells which would proliferate. And we call them
ES cells. It's a total tissue culture artifact of those cells
which can survive under the harsh tissue culture conditions. So
I would argue ES cells have no counterpart in the normal animal
model. They are a real tissue culture artifact, although a very
Now the same occurs, of course, with these blastocysts. The efficiency
is lower. But once you go to the selection for the survivors, they're
exactly—I think in order to survive, they have to express the
set of 70 genes which we call the embryonic genes, which are important
for the early developmental stages. So the point of this slide
is really that this selection process, which selects just for the
fastest growing cell, erases the memory to where the ES cell came
from. And indeed, when you transplant these cells derived from
a B- or T-cell or from a neuron or from a zygote, they have identical
properties. They form normal chimeras, and they in vitro
I want to summarize this. An ES cell derivation selects for survivors.
And I would argue survival depends on an ES default epigenetic state
which needs these Oct-4 like genes being on. Selection process
erases the epigenetic memory of the donor nucleus, and the cloned
and the fertilized cells form normal chimeras. So the potential
of ES cells derived from an in vitro fertilized embryo and
from a cloned embryo is identical by all measures we can do.
So I would conclude, then, it's unlikely, if not impossible,
to create a normal individual by nuclear cloning. The problem of
reprogramming may not be solvable for the foreseeable future because
of these principle barriers, but ES cells derived from clone embryos
have the same potential for tissue repair as those from the fertilized
So I think one of the key concerns I can see of this committee
is that the derivation of embryonic stem cells by nuclear cloning
necessitates the destruction of potential human life. I think that's
a major concern. And if I just compare now, to my opinion, the
difference between a fertilized and a cloned embryo. The fertilized
embryo is created by conception. It's genetically unique.
There's a high potential it will develop to a normal baby.
The cloned embryo of course has no conception, no new genetic combination.
It is really the product of a laboratory-assisted technique. Sloppily,
we could say it's a laboratory artifact. But most importantly,
it has little or no potential to ever develop to a normal baby.
So I think the embryo, the cloned embryo, lacks essential qualities
of the normal embryo, which on this sort of maybe summarizing my
thoughts is on this slide where I think there are really three possible
fates for a cloned or for fertilized embryo. We fertilize one of
these leftover embryos, hundreds of thousands in the clinics. They
can be disposed of, they can be implanted to form a normal baby
with a high probability, or they can generate normal ES cells.
The cloned embryo, three fates. It can be disposed of. It can
make normal ES cells, as I have argued, but it cannot make with
any acceptability efficiency a baby, not even a normal one.
So if this is accepted, instead of disposing these leftover embryos
and use them for normal ES cells, which could be used for research,
as we heard. This, of course, to my opinion generates an ethical
problem because you destroy potential human life. I think this—so if this is acceptable to some, I think it should be ethically
less problematic, because in this case I think you don't have
the potential to form, within acceptable possibility, a normal baby.
So from the biological point of view, I think the derivation of
embryonic stem cells by nuclear cloning develops the structure of
an embryo that lacks the potential to develop into a normal being
with any acceptable efficiency or predictability.
I was asked—and I want to close with my final slide, which really
follows what John said—what are the potential applications of
cloned embryonic stem cells derived from a cloned embryo. I think
we talked about therapy. Clearly, I don't have to dwell on
this. I'm not sure whether this will be really generally available
technology. It maybe too expensive. I don't know. And I don't
know how fast we could solve the technical problems of adapting
this to human medicine.
However, I think this is the more important point, which I think
John emphasized. I think it allows us to derive genetically identical
cells from patients with multigenic diseases, such as Parkinson's,
Alzheimer's, ALS, diabetes. And we can now use this system
in vitro to validate the cellular defects of these complex
diseases. So compare ES cells derived from a healthy individual
with those from such a patient. I think there's enormous potential
here if we find a difference, to find out why that is so, and is
it a screen for potential therapeutics, just in the culture dish.
And as John emphasized, such diseases cannot be studied in animal
models, because there are no animal models for those diseases.
But you can really make those in the culture if you just use a cell
of the patients.
So I think this is a very important, maybe the most important
driving issue, to my opinion, to use this technology. Thank you.
CHAIRMAN KASS: Thank you very much. Dr. Prentice. While
we're waiting, Dr. Jaenisch, would you mind just taking one
question of just factual information so that we don't lose a
lot of time. This last point about the models for multigenetic
diseases. If someone were to say the same kind of models might
be available through, for example, the multipotent progenitor cells
that Dr. Verfaillie has, where you could go into the patient and
get out progenitor cells. You might not get all the tissues, but
if those cells could reliably be differentiated into islet cells,
couldn't you do the studies on the pathogenesis of—in other
words, couldn't one, if one had stable and reliable adult stem
cell populations, wouldn't you have the same access to the cells
from the multigenetic diseases?
DR. JAENISCH: Yes. I think if this would work, if this
would efficiently work, I think absolutely I agree with you. I
think the problem as I see it now is that, indeed, working with
adult stem cells is very difficult. And the major problem is that
we lack the ability in most cases to, for example, propagate them
in the undifferentiated state. For example, with bone marrow stem
cells, although they're known for 30 years, this has not been
accomplished. They are very useful for therapeutic applications
because they select themselves in the patient, but not easily for
this type of research, what you want to do.
So I think there might, at some point, if we learn that in this
young field, I agree with you. Then I think these cells would,
if you could differentiate them to dopaminergic neurons in a way
that we can do for now from ES cells, as Ron McKay has shown, yes
I think that would be useful.
CHAIRMAN KASS: Thank you. Dr. Prentice, you're all
DR. PRENTICE: Yes.
CHAIRMAN KASS: Good.
DR. PRENTICE: I think we're ready to roll. Thank you,
Mr. Chairman. I apologize for the delay up here. I was fighting
off a fever from a respiratory virus, so I probably was in a time
One of the main goals of stem cell research, as you're all
aware, is the idea of regenerative medicine with stem cells, the
idea that you might be able to take a stem cell from some source,
for example here from bone marrow, inject that into an area of damaged
tissue in the patient and regenerate or replace the damaged tissue.
Now, the interesting, confusing thing about adult stem cells has
been that it defies what for years we have thought of as the normal
developmental paradigm, that as we develop from the blastocyst stage
here, that cells follow one of these main developmental trees.
And as a cell would become more and more developed and more and
more differentiated, they would end up out here on the tip of a
branch and not be able to back away from that particular tip; not
be able to back down and take a separate branch, whether it were
a nearby branch or one of these main branches.
A lot of evidence now suggests, however, that adult stem cells,
at least some of them, can actually move from branch to branch.
Now exactly how they're doing that, the mechanism involved in
these types of differentiation or different branches of tissues,
is still unknown.
There's several questions related to adult stem cells, and
in fact many of these have to do with the same questions related
to embryonic cells. What is their actual identity? How could you
identify a particular adult stem cell? What is their actual tissue
source? How do they form these other cell and tissue types? Do
they form actual functional cell and tissue types, as we look at
these organs and tissues? And what's the specific mechanism
They do seem to have some unique characteristics, such as a homing
phenomenon, where they tend to home in on damaged tissue. And most
of the results that have been seen in terms of an adult stem cell
differentiating into another cell type or tissue type seem to be
tied primarily with injury to a tissue or organ. Usually you do
not see these types of changes in the experiments taking place unless
there is some sort of damage. The other thing would be what type
of cellular interactions and signalling within the target tissue
might trigger some of these differentiative events.
Now, one way to try and identify a stem cell is with various markers,
and I discuss this at some length in the paper. Typically what
people have tried to do is eliminate what are termed "lineage
markers" for particular blood cells. Some people pick a particular
marker called CD34, which has been associated with bone marrow.
But others tend to eschew that type of marker. And I mentioned
other markers in the paper that might indicate that an adult cell
is a stem cell. The CD133 marker, or the c-kit marker.
Several people have begun to undertake studies of the gene expression
within adult stem cells, and compare them to embryonic cells to
see if there is a commonality in terms of all of the various genes
that might be expressed that identify a cell as a stem cell. One
of the problems that's been faced, though, is that these markers
tend to change over time. This has been seen in several studies
where actually a marker such as CD34, that might have been a chosen
marker to identify a stem cell, actually might not be expressed
at a later time with that cell. And then expression of that gene
would re-occur, even within the same cell.
So it may be difficult to actually identify one particular stem
cell based on some of these markers, simply because of changes in
gene expressions as the cells undergo changes in environment, as
they may undergo changes in their isolation conditions, and what
Theise and Krause simply have called the uncertainty principle,
equivalent to the Heisenberg uncertainty principle. You might at
one point be able to isolate a cell with a particular set of markers
or milieu of gene expression and say this is a stem cell, but then
as you would try to put it into different conditions, or isolate
it from different tissues, those particular markers might change,
or might be different.
So it may actually be very difficult with adult stem cells to
particularly isolate a particular cell and say, yes, this is an
adult stem cell. It may be more a matter of the particular context
of the cell: the tissue they're derived from, the isolation
conditions used, or the tissues that they're put back into that
may determine their functionality.
In terms of differentiation mechanisms, cell fusion appears to
be one particular mechanism by which an adult stem cell may change
into another tissue type. Now there were some in vitro experiments
done approximately a year and a half ago that indicated this possibility,
but the in vitro experiments were unable to verify whether
this was actually a particular mechanism that might be used by an
adult stem cell. More recently, two papers have verified in mouse
systems that bone marrow stem cells, at least in these experiments,
did fuse with hepatocytes and take on that differentiated morphology
and function, even in terms of repairing liver damage. So this
is obviously one possible mechanism with some experimental evidence.
Another possibility might be that a cell would de-differentiate,
actually back down one of those branches and go up another branch.
This would involve changes in gene expression to an earlier state,
or more primitive type of cell, and then re-expressing specific
genes particular to the tissue into which the cell was placed.
Another possibility is what's termed "trans-differentiation"
in which the cell apparently is not backing down a branch, but instead
simply changing its gene expression so that it now conforms to whatever
tissue or cell population that it's within.
In terms of sources of adult stem cells, not just tissues, but
how did they get there? How did an adult stem cell get into bone
marrow, or brain, or liver? One theory is that some of these cells
may be leftover primitive stem cells, perhaps embryonic stem cells,
a few that are kept around, maintained in a state so that then they
can differentiate into various types of tissues.
Another proposal is that there may be a universal stem cell, an
adult stem cell, not a primitive cell, but one more geared towards
maintenance and repair functions in the adult body. It may arise
in one or more tissues, and then may disperse into other tissues.
Other possibilities are that there are particular tissue stem
cells, multipotent, not able to perhaps form all tissues of the
adult body, but a limited subset, and that they would be resident
within a few of these tissues.
And then I mentioned here transient stem cells, cells that are
just sort of passing through. And there is the possibility that,
especially if we're looking at this idea of a universal stem
cell, or even a tissue stem cell that can migrate in some cases,
that it may end up passing through a tissue, so you will isolate
what appears to be a stem cell from a particular tissue. But it
didn't arise there. It just happened to be passing through
via the circulatory system.
I might mention, too, that some of the other challenges for adult
stem cells are very similar to some of the challenges for embryonic
cells. For example, standardization. There essentially is no standardization
in terms of adult stem cell isolation, propagation, differentiation
into other tissues at this point. Long term culture, with a few
exceptions, has not been attained. Catherine Verfaillie's MAPC
cells from bone marrow do appear capable of long-term proliferation
in culture. And there are a few other examples, very limited, that
I point out in the paper. But being able to keep these cells growing
in culture for a long period of time has been difficult up to this
The idea of safety, of course, is always an issue when we're
going to be dealing with patients. Do these cells form tumors?
Do they differentiate abnormally so that you don't get the correct
differentiation in the place, in the time, in the tissue that you
need? So all of these need to be faced.
Now this idea of a potential universal or at least tissue stem
cell that can migrate primarily has been put forth by Helen Blau
of Stanford, the idea being that perhaps the cell is in the bone
marrow, but then can go into the circulation and then exit the circulation
into another tissue. Again, most of the studies have seen this
type of phenomenon in response to an injury, not in response to
the normal physiological systems.
How would we then define a stem cell, especially an adult stem
cell, if we have all of this problem with changeability of markers,
and lack of standardization? Moore and Quesenberry have proposed
a couple of simple guidelines. Now obviously, any stem cell has
the ability to continue to replicate and maintain a population of
cells. And then in response to some signal, to differentiate into
one or more potential tissue types.
An initial first cut at how to identify a stem cell in the adult
would be, can it take on a different morphology in a tissue in which
you place it? Can it take on some of the differentiated cell markers
that you would see in that tissue? Now this is a bare minimum,
and it doesn't really mean necessarily that the cell is functional,
that it can participate in tissue repair, or perhaps even that it
has truly differentiated into that specific cell type.
Supplementing that, if you could demonstrate functional activity,
and actual integration into a tissue, this would be a much better
marker of a cell as a stem cell. And certainly, if in an injury
situation, in animal model or even in human, you could demonstrate
a physiological improvement. That would be a good indication that
you were seeing this sort of stem cell differentiation and repair.
Now, the next set of slides, what I've tried to do is just
condense from that rather difficult mass to read of all of these
various tissue sources, and cell types that they can turn into,
and so on. I won't spend a great deal of time on these. I'll
just try and point out some particular points on each slide.
With mesenchymal or stromal stem cells from the bone marrow, we
need to keep in mind that bone marrow is a mixed population. We've
known about hematopoietic stem cells from bone marrow since the
1960s. These are well-studied and used clinically. The hematopoietic
stem cell is differentiating primarily, at least, into blood cells.
Another stem cell that's in the bone marrow is this mesenchymal,
or stromal, cell. As I mentioned in the paper, one way that this
type of cell is actually isolated is perhaps first by markers, but
then also simply by its ability to form attached adherent layers
in the culture dish, once you're putting them into the dish
in the lab.
Now, there have been various studies, and I've just tried
to collect a number of them here, showing potential differentiation
into different tissue types, either in the lab dish, or in the animal.
Again, the way they're following this in the animal is primarily
to use a genetic or a fluorescent marker, follow the cell, see what
tissue it ends up in, and look at its morphology, and then also
potentially try to analyze functionality and integration into the
In various disease models, some of these cells have been used.
For example, in models of stroke. And I want to point out again
what Dr. Gearhart said. Animal models are only an approximation.
This does not tell us the actual situation that we might encounter
in human beings. But at this point, it's the best we can do.
We can see if these cells might participate in cell and tissue repair.
Using rat or human cells, there has been some therapeutic benefit
in some of these disease models in stroke with animals. With demyelinated
Intravenous injection, interestingly enough, of mesenchymal stem
cells in the mouse has shown some remyelinization in spinal cord
injury. Adult stem cells are not alone in terms of this ability.
Other cell types have shown this. And this does not necessarily
mean that you've totally corrected the condition. It simply
means that you're able to provide some sort of functional recovering
of a bare nerve.
Another interesting point about this particular study was that
they did not see a large incorporation of these cells into the spinal
cord tissue. So what benefit they did see did not appear to be
due primarily to the adult stem cells actually differentiating into
the myelinating cells here. Instead, what appeared to be happening
was the presence of those cells caused a signal for the endogenous
cells to start to re-grow and repair the damage.
In fact, Dr. Gearhart's lab just recently published a paper
showing that embryonic germ cells showed in a particular mouse model
the same type of repair phenomena. The mice did show therapeutic
benefit. But what they found was that the cells did not, to a large
extent, integrate in and participate in the repair. Instead, it
seemed to be a signalling phenomenon. And I'll point this out
in a couple of other examples along the way.
There has been one clinical trial that I could find in the literature
where 11 patients were treated for Hurler syndrome, or metachromatic
leukodystrophy using allogeneic, and not from the original patient,
but from donor mesenchymal cells. Now, the study noted that four
out of those eleven patients showed a small increase in nerve conductance.
One of the problems with these diseases is a decrease in nerve conductance.
There are other neurological effects, and so on. So small increase.
And in terms of loss of bone mass, all 11 patients seemed to maintain,
at least for a certain time. But in terms of any of the other symptoms
in these patients, there was no effect of the adult stem cells.
Very early study.
Using mixed population of bone marrow stem cells. So in this
case, and again, this lack of identifiability for a particular adult
stem cell. What many people have done is simply take raw bone marrow
and either in culture try to get various differentiations, or putting
back into the animal in various disease models. And they've
been tried in quite a number of disease models showing some therapeutic
benefit. In this particular case, again in stroke, intravenous
injection of this mixed population of cells. Again, no attempt
to purify the cells, or identify the particular cell that's
being used for the treatment of the animal model. But the cells
in this case did home in on the damage. Again, this interesting
phenomenon that we've seen.
Various other conditions. There was an interesting study relating
to cardiac damage, where cells from mice were actually put into
rats. And what they noticed was over a fairly extended period of
time there did not appear to be immune rejection of these transplanted
cells, even though they were even from a different species.
Now, it's only one study. Does that mean that these particular
cells might be immune privileged and not subject to rejection?
Only more experiments will tell. Recent study indicated that with
diabetes, in a mouse model again, where the pancreas was damaged
to induce loss of insulin secretion, bone marrow stem cells injected
into the animal could regenerate insulin-secreting cells. As in
the other example I mentioned before, there was very low incorporation
of the actual bone marrow stem cells identified as forming insulin-secreting
cells. Instead, what they seemed to be doing in this instance was
again secreting particular growth or stimulatory factors to stimulate
the endogenous cells to re-form insulin-secreting cells.
And there have been several clinical trials and publications,
peer reviewed publications, from those trials, primarily with cardiac
damage. There have been several groups in Germany and a group in
Japan that have published their results. Again, very, very limited
numbers of patients, we must keep in mind. And what they're
seeing, though, is some physiological improvement in the ejection
fraction from the heart, and when they actually examine some of
the damage that seems to be being repaired. One Japanese group
has reported growth of blood vessels in limbs that were threatened
to be lost by gangrene.
Peripheral blood. Probably we should just skip this category,
in a sense, because they seem to be very similar to the bone marrow
stem cells. And it makes a lot of sense that bone marrow stem cells
could actually enter the circulation. The interesting thing is
that two groups recently have isolated peripheral blood stem cells
from human blood. One group in particular noticed that the cells
that they isolated showed high levels of Oct-4 expression, this
gene that's associated with pluripotent status of a stem cell.
In a couple of animal disease models for stroke and cardiac damage,
and I put this under peripheral blood because what the workers did
was not remove adult stem cells and then re-implant them in the
animal. They injected growth factors to stimulate mobilization
of the stem cells, most likely from bone marrow. And these cells
appeared to provide a therapeutic benefit in these conditions.
And we'll come back and talk a little bit more about mobilization
Neural stem cells. I certainly was taught in graduate school
that you start with as many brain cells as you're ever going
to have and it's downhill from there. And on my campus, it's
downhill even faster on a good Friday or Saturday night. But within
the last decade or so, the discovery that there are these stem cells
within neural tissue that can at least regenerate all three main
neuronal types, and perhaps some other tissue types as well.
Interestingly enough, they have been used in some animal disease
models to show some therapeutic benefit. And there have been some
clinical trials, very early. There is one Parkinson's patient
who was treated with his own neuronal stem cells. The cells were
removed, grown in culture for awhile, and placed back just into
one side of the brain. He did see a benefit. He did see improvement.
In terms of spinal cord injury, there are no peer reviewed publications
in terms of actual human treatment. But there have been reports
that at least one group in Portugal are beginning these types of
treatments on patients for spinal cord injury.
Dr. Gearhart mentioned the embryonic carcinoma cells. And there's
one group of this cell termed usually "hNT" that you might
say has been tamed in culture. This particular line that they've
come across primarily forms neuronal-type cells in culture, and,
as he mentioned, it's being used in a beginning clinical trial
Muscle stem cell can regenerate itself. It has its own stem cell,
primarily dedicated to forming more muscle. But there's indication
that it does have, perhaps, another stem cell present in that population.
Debate as to whether those cells can actually form blood, or if
that was one of these so-called transient cells moving through.
The potential though, perhaps, to form bone from this type of cell,
and these muscle stem cells have been used in some disease models,
primarily directed back towards muscle. Very interesting one that
I came across was to re-form bladder muscle, or at least reinforce
bladder muscle, in a rat model of incontinence. And there's
been one published paper regarding a clinical trial using muscle
stem cells, one patient, in France, where the patient did seem to
receive some therapeutic benefit from the muscle stem cells.
Liver stem cells and pancreatic stem cells. Pancreas and the
liver derive from the same embryonic primordia. And so it might
not be surprising that you tend to see liver able to form pancreas,
or pancreas able to form liver. The interesting thing here is that
genetically engineered liver stem cells, simply by adding and activating
a gene, have helped an animal model induce diabetes in these animals.
The pancreatic stem cells, likewise, have shown the similar ability.
Corneal limbal stem cells are a very limited subset. And what
we might in one sense call a unipotent stem cell, at least for the
most part, other than one or two little reports, so that they would
re-form only the tissue in which they're located. They have
been used, however, in corneal transplants in a number of countries.
And there are several published reports on this. In vitro,
one group was able to get them to form neuronal cells in culture,
or what looked like neuronal cells.
Cartilage, again, a very limited sort of application, but clinical
trials for articular cartilage transplants. And one group has used
them in an attempt to treat children with osteogenesis imperfecta.
Umbilical cord blood, not a great deal of research has been done,
but there are a growing number of publications. The interesting
thing here is, in terms of the disease models, stroke, spinal cord
injury, and one particular model of ALS, the cells are delivered
intravenously. So again, this potential to home in on areas of
And then just a list of a number of other tissues here where some
type of adult stem cell has been identified, or potentially identified.
And I want to point out the adipose, or fat-derived stem cells.
And I use "derived" because it's very likely this
is one of those transient stem cells moving through via the circulatory
system. Interestingly enough, though, it provides a reliable, readily
available source, you might say, from liposuction fat. And one
group has been able to see neuronal differentiation in the lab;
not in any animal models yet.
Now, in terms of activation and mobilizing these adult stem cells,
this may be actually one of the best directions to go in terms of
potential clinical treatments. Simply because it may not be the
stem cells themselves that participate in the repair of the damage.
And in that case, maybe what we need to do is simply activate the
cells so that they can deliver the signal to the damaged tissue.
And again, there may be various states within the bone marrow as
an example of a tissue where you might activate, where the cell
could be activated and then mobilized into the circulatory system,
and then reach the particular tissue to differentiate.
Again, there were two animal models that had been used, one for
cardiac damage, and one for stroke, where there was an indication
that this might be a potential avenue of research. Rather than
removing the cells, having to culture them, isolate them, identify
them, put them back, simply being able to mobilize the cells within
Another possibility is genetic engineering. And in terms of actual
animal models of disease, genetically engineered adult stem cells
have been used in a few examples in the publications. In lung,
that one particular reference indicated that there was a large contribution
of the adult cells to lung tissue. Again, a damaged tissue system.
Clinical trials, not directly, you might say, repairing a damaged
tissue. But in Severe Combined Immuno-Deficiency Syndrome, bubble
boy syndrome, several infants have been treated using genetically
modified bone marrow stem cells from the patient. They seem to
have been, quote, "cured," and I use that word advisedly,
of the disease. There have been reports that at least two, and
potentially a third infant, has come down with leukemia as a result
of that treatment, most likely due to the viral vector that was
used to insert the gene into those cells. Because the gene is not
targeted in those instances. Instead, it may go into an area where
it can activate oncogenic genes within the cells.
In terms of delivering the signal, this might be an even better
use of genetic engineering in adult stem cells. And there have
been a few published examples, potentially for use in animal models
with tumor therapy, to alleviate some of the symptoms in the model
of Niemann-Pick disease for increasing bone healing, and in a Parkinson's
model, where the cells were not participating to any great extent
in repair of the tissue. Instead, they're delivering a signal
to the endogenous cells there, and stimulating them.
And in terms of that then, perhaps the best avenue eventually
to pursue might be trying to isolate what these factors are, so
that no stem cells would be needed at all. Instead, if you can
identify those factors to stimulate regeneration within the tissue,
they could be delivered directly.
And there are a few published examples of this, one in which there
was some axon regrowth, some, in a rat spinal cord injury model
by injecting a couple of different factors. One mouse Parkinson's
model study infused directly a factor into the brain of the mouse
model and saw a therapeutic benefit. Another publication lists
a different factor in terms of a model of chronic kidney damage.
And one clinical trial where five patients, five Parkinson's
patients were treated with infusion of this glial-derived neurotrophic
factor, they did see some improvement. Not a cure, and it may be
that that's not the sole signal that's needed, or the correct
signal. But it did seem to stimulate some physiological improvement
in the patients.
CHAIRMAN KASS: For a young field, there's a lot going
on. And we have also talked our way late into the hour. Let me
make a suggestion, if I might. Technically, this session was to
run till 3:30, with the next to start at a quarter of 4:00. I'm
going to take the liberty of suggesting we spend a half an hour
here at least with our visitors to get some of our questions asked.
And if Lori Andrews will forgive us, we'll be pushing back the
start of the next session, just so we can get the questions asked.
Michael Sandel, then Jim Wilson.
DR. SANDEL: I learned a great deal from all of these presentations,
as I suspect we all did. I was especially struck by one feature
of Dr. Jaenisch's presentation, and I'd like to put this
as an observation which actually will end with a question.
The subtitle of the paper and of the talk, Dr. Jaenisch, and my
Council colleagues will appreciate this perhaps more than our visitors.
The subtitle could well have been, "The Vindication of Paul
McHugh." Because when we were discussing the biology of cloning,
almost everyone here shared an assumption, regardless of the positions
that we took on the ethical issues. Almost everyone took it for
granted that biologically the fertilized embryos and cloned embryos
were essentially the same, and the moral argument proceeded from
that assumption. With the exception of one person, Paul McHugh.
And he introduced a distinction, a distinction, as he put it,
between a zygote and a "clonote", by which he meant the
cloned embryo, the product of cloning. And for insisting on this
distinction, poor Paul suffered heaps of ridicule and abuse, and,
at best, a kind of bemused chuckling, by those of us around the
table. And he argued that there's a difference, a biological
difference, with a possible ethical significance between a zygote
and a clonote, between a fertilized embryo and an artifact created
in the lab. And he was told that this is an eccentric position.
He was told that this is an off the wall distinction, that it had
no basis in biology or science.
And in a memorable exchange with our colleague Charles, who unfortunately
isn't here for the benefit of this discussion, he was asked,
"Well, but what if a cloned embryo actually came to term?
Isn't Dolly a sheep?" And Paul's answer was, "Well,
a sick sheep."
And then we read, Dr. Jaenisch, in your paper that a cloned human
embryo would have little, if any, potential to develop into a normal
human being, that it lacks essential attributes that characterize
the beginning of normal human life. And you explained that it has
to do with faulty reprogramming after nuclear transfer.
Now, I don't know whether that's true, whether it's
generally accepted, or controversial in the scientific community.
But whatever the case, it seems that what we regarded here as an
eccentric, off the wall suggestion, that there is a biologically
significant difference between a fertilized embryo and a cloned
embryo, now, if I understand you correctly and here's my question,
now, you're telling us that far from being eccentric or off
the wall, Paul's position may be true.
DR. JAENISCH: Well, I think that was the whole purpose
of going through these arguments for me. To argue from a biological
point of view, it is a continuum. We have to struggle with that.
And we have, for the early, most abnormal stage, we have good markers.
Death is a terrific marker. Then we have molecular markers after
birth, or we have aging as another marker.
So these are concrete markers, and we know these animals are not
normal. So when it is claimed that cows on the field are normal,
which has been published by ACT. They looked at their one- to four-year-old
cows sit on the fields and eat and give milk. They argued, well,
they must be normal. So they checked that. And they checked it
by checking if it had a heart. Yes, it had a heart. A liver, yes
they had a liver, and that serum, even. So they concluded they
must be normal and publish this in "Science".
Now, this of course is a very superficial way of looking at it.
And very interesting. Once this paper came out, six months later,
two of these healthy cows, one got multiple tumors, the other one
generalized seizures. Now this is not normal. You wait, and these
things come out.
The point I was trying to make is that these are stochastic problems.
There is not a key master gene which, if this is right then the
embryo's right. No. The whole genome is stochastically, correctly
reprogrammed or not. And what's amazing to me is that mammalian
development is so regulative that it can tolerate so much gene disregulation
and still get an animal to birth and beyond. But you pay for that.
You pay with subtle or less subtle, more or less subtle problems.
And of course, because it's stochastic, you will arrive at
the end at some animals which live a normal life span. And by all
our available tests they might be normal, or you might not see it.
If you look at the primate, or in the human, we might see this because
we could look at some much more subtle phenotypes.
So I think the point I was trying to make is if you want to use
this as a technique to generate human beings, there's no predictability.
You cannot do this. And for me, it doesn't matter with one
percent or two maybe something normal, because you can't predict
So I totally agree with what your thought was.
CHAIRMAN KASS: Jim Wilson.
PROF. WILSON: Michael asked the same question I was going
to ask using virtually the same words. I guess it's the Harvard
But I want to just press a bit more. An oocyte is not potential
human life because it's an unfertilized egg. You can enucleate
it and put in a new nucleus, and now it is a clone. But according
to your account, a cloned embryo has little, if any, potential to
ever develop into a normal human being.
If you would step back, then, from your biological role and adopt
a somewhat more philosophical role, would you say that a clone,
therefore, is not potential human life?
DR. JAENISCH: I would argue it's not potential normal
PROF. WILSON: Thank you.
DR. JAENISCH: I think that's the only thing I can
CHAIRMAN KASS: And Dolly still is a sheep?
DR. JAENISCH: Still was a sick sheep, was a very sick
sheep, and which died, of course, predictably very early.
CHAIRMAN KASS: Janet Rowley.
DR. ROWLEY: Well, I, too, learned a great deal this afternoon.
I want to make two comments, and then I have a question for Rudy.
The first comment is that I think it is most unfortunate that
we're prevented from asking our experts, and particularly John
Gearhart, about the status of funding that supports this very critical
area of science.
The second comment is also directed to John. And I was struck
by the specific chromosome abnormalities that you identified in
your human embryonic cells, because you probably know that an isochromosome
for 12p is the most common abnormality in human testicular carcinoma.
So it's astonishing that this is the abnormality that you
see in these cells, and I'm sure it is extraordinarily important.
The question that I have for Rudy—I was surprised when you said
that imprinted genes are only important in embryogenesis, because
clearly there are a number of human conditions.
Now, you could say, well, human condition and the malformations
occur because of embryogenesis, but there is also this whole question
of imprinting and its relationship to carcinogenesis. So I'd
like you to expand on that.
DR. JAENISCH: This is a very important point you're
bringing up, so I was probably a little bit generalized. So I think
most imprinting, as we know, have—so that could be among these
things that could be the most interesting, because it's IGF-2,
which is a loose cannon.
IGF-2 is a tumor, an oncogene essentially, so this would be available
for cloning, because IGF-2 is very easily dysregulated. And I would
worry if you have both copies active, because that's pretumor
status. So you want to screen for this.
But the other genes—for all of the other genes, for example,
an Angelman Syndrome probably really is—there are really problems
of fetal development. So these genes and fetal development—I
could go further into this, because it's really growth of the
fetus, where these genes appear to play the most important role.
And there's really no evidence, with the exception of IGF-2,
maybe KIP-257, which is a tumor suppressor gene, those are the ones
you would worry about. So I think we know only two of those, and
those you would test for I think.
So I agree, which makes this a little bit more restrictive when
I say it.
DR. ROWLEY: Right. And there probably are others that
we don't know about, but—
DR. JAENISCH: There might be others which we don't
DR. ROWLEY: —you're saying that they would—
DR. JAENISCH: Yes.
DR. ROWLEY: —just all be tested for.
DR. JAENISCH: Those two we know for sure, but others we
know they don't play a role.
CHAIRMAN KASS: I have myself next, if I might. A question
for Dr. Prentice, and then for the group as a whole. In Dr. Gearhart's
presentation, he gave a great deal of emphasis to the importance
of the characterization of these cells, the reproducibility, all
of these various things.
You had to cover an enormous amount of material in the number
of papers, and you also in your presentation alluded to the problem
of characterization, though I gather many of these studies seem
to be one-shot efforts of throwing some things in and trying to
see what their clinical effect might be.
And as I've read the debates, a lot of people wonder, are
these really stem cells? How well characterized are they? So could
you say something about what criteria should be used for these stem
What criteria would they have to meet before you could satisfy
the critics that these are, in fact, fit to be called human stem
cells, and fit, therefore, for controlled studies on the kinds of
things that you're talking about here?
And kind of a subordinate question, what's the best characterized
of, and the most versatile of, these adult stem cell preparations?
Are there any that at the moment meet the criteria that you yourself
DR. PRENTICE: In answer to your first question, it's
difficult, obviously, because as you point out many of these studies
have been sort of one-shot, especially if you look at the mixed
bone marrow population studies where they are simply taking a preparation,
putting it back into a diseased animal or a diseased model, and
trying to come up with some therapeutic benefit.
You know, obviously, what you'd like to be able to do is to
isolate the cell or the population, keeping in mind that caveat
that the population dynamics may be an important facet of this.
But let's say there's one particular stem cell that you're
after. Supposedly, if you saw that sort of therapeutic benefit
from the mixed population, you should be able to go in and simply
take cuts and eventually isolate that one cell, and then be able
to grow it in culture.
In terms of the overall markers, that's been very difficult,
and there have been a few studies looking at the gene expression
common to neural stem cells, bone marrow stem cells, embryonic stem
cells. And there are a subset of those genes that are common to
all of those cells.
Now, what you would hope to be able to do, then, is to go look
at each of those genes and start using those as guides to come up
with a marker for any stem cell that would be targeted for the types
of differentiation models that we'd like to see, and eventually
CHAIRMAN KASS: And with respect to the second question,
what is the best characterized and most versatile adult stem cell
preparation at the moment? I mean, can you—is there—
DR. PRENTICE: Probably Dr. Verfaillie's MAPC cells,
and I can't speak for her, but those do show many of these characteristics
that ideally you would like—long-term growth in culture, clonogenic
growth, and then differentiation. They've been used in a couple
of different animal models of disease to see some sort of benefit.
One of the mesenchymal cell—stem cell lines, if I remember correctly—and there are a lot of things to sort back through there—had
been kept in culture, I believe, for about two years. There was
another cell line that—I think it was also a bone marrow derived
cell line, but it had been genetically engineered to contain telomerase
or additional telomerase gene. And that had been kept in long-term
I believe also one of the neural stem cell lines has been kept
in long-term culture, and actually Dr. Gearhart might be able to
comment more on that.
CHAIRMAN KASS: Okay. Thank you.
And I have a general question to all of you, if I might. Since
we sent you some general questions that we were hoping to get some
help on, to go from where we are now to actually having reliable
tissue grafting therapeutic applications in humans, since there's
a fair amount of speculation out there—and, Dr. Gearhart, you
did emphasize how early we are in the process and the emphasis,
really, on learning of the basic biology.
But the public out there is more—much more interested in, you
know, where are the cures? Could you address that particular question
of the various sort of stages, and where we are, and what kind of—how one should talk about these things responsibly in relation
to the widespread hopes for remedies for which the patient groups
are, quite understandably, clamoring?
DR. GEARHART: The answer to that question is a very complex
one for me. I could run through a number of cell lineages with
respect to the human embryonic and stem cell and germ cell lines,
where we are in the process, but it's the other end that I want
The most difficult talks that I give anymore are those before
patient-based groups where the hopes are high. They come with expectations
that far exceed, I think, where we are. And this is the result
that I think that early on in the field, as you mentioned, the hype
was there when circumspection should have been there. And I don't
think circumspection is a retreat from promise.
The reality is when you get into the laboratory, and you are working
on cell lines, it seems the more progress you make, the further
away the goal is as we learn. And we can show I think remarkable
things in our animal models of putting in cells, and these are almost—they bear some resemblance to the adult stem cell story where
you try a lot of things empirically, you get a response of some
kind, and you're trying to figure out what it is, why.
But when you superimpose on top of the biology issues of trying
to control these cells in a dish, and trying to expand them in a
dish to characterize them, to show that they're authentic, that
they do work, the issues of safety, which I think we all are in
agreement have to be foremost in this, that the time frame is years
away before I think any of this will be realized in a clinic.
I think there's misunderstanding, that if all of a sudden
you have an insulin-producing cell in a laboratory, which we do,
and we're—as other labs have as well—that most people think
this is the answer. You just put insulin-producing cells into a
diabetic. And as you know, this doesn't solve the underlying
We can—so I don't know how to answer the question that you
ask. I mean, I'm certainly optimistic. I think it's going
to be a direct function, and I'll come back to Janet's question
that she can't ask or isn't supposed to ask, and that is
that obviously more funding into this area—and I know that this
isn't a congressional committee of any kind, where we're
seeking funding, but obviously the progress is directly proportional
to the number of good investigators you have working on projects.
And in this field, we've had a bottleneck, first from the
number of lines that could be used, now the NIH funding I'm
sure is going to be more robust over time, but this remains not
only I think a financial issue but also almost a psychological one—for students getting involved, for post-docs getting involved,
for young faculty getting involved in the field, where we don't
know what the future will be, to be honest with you, in this country.
We see our colleagues around the world in different laboratories
I think making really substantial progress here. And so I'll
get off this part of it, but I think this is an important issue,
at least with the embryonic and fetal sources of cells. And I hope
we'd get over that.
CHAIRMAN KASS: Thank you.
I have Gil Meilaender, Bill Hurlbut, Janet. Gil.
PROF. MEILAENDER: Yes. Dr. Jaenisch, I want to come back to
where Michael Sandel started us, because I'm just not certain
that I'm clear on something. And what I want to know is, you
know, it's always hard to know when one's speaking as a
scientist and when one's speaking as something else.
But as a scientist, there are various kinds of chromosomal or
genetic defects that a fetus might have, the result of which would
be that it might die immediately after birth or very soon after
that. How would you scientifically describe those fetuses? Would
you say that they have—that they also do not have the potential
to be a normal human being?
I'm trying to get away from the stem cell issue for a moment.
Would the same characterization be the appropriate one there?
DR. JAENISCH: This a really a question which is not a
scientific question, then, right? An embryo which has a trisomy
number, whatever, large chromosomes will die very early, maybe before
implantation. So that has this embryo which is afflicted with this
Has that a normal potential development to a human being? No.
PROF. MEILAENDER: So it falls into the same—so, in other words,
this is a rather large category of embryos and fetuses that—about
which we would say this. We're not only talking about cloned
DR. JAENISCH: Well, I think that—
PROF. MEILAENDER: The same characterization—
DR. JAENISCH: Yes, I—
PROF. MEILAENDER: —would fit.
DR. JAENISCH: I think the difference, from my point of
view, if a chromosomal abnormality occurs after normal fertilization,
something which is totally unpredictable and is assessed, that is
what fate is to genetic game. With cloning, I think it's predictable,
and that's the difference. It is predictable to a large extent,
so I think that would make the difference.
One is spontaneous, unpredictable, it's a risk of normal reproduction.
The other one is predictable. Otherwise, I really can't—I
think this for me is a major difference. You predictably generate
a human being which has no chance to become normal with any acceptable
PROF. MEILAENDER: I think I'll stop there.
CHAIRMAN KASS: Bill Hurlbut, and then Janet.
DR. HURLBUT: Did I hear you correctly say you find destruction
of IVF embryos morally troubling, but cloned embryos for embryonic
stem cell use less troubling?
DR. JAENISCH: No. Well, I said that if you accept that
instead of throwing away an in vitro fertilized embryo which
cannot be implanted because nobody wants it, instead of throwing
this embryo away generating to an embryonic stem cell, I mean, people
accept this. I find this acceptable. But it does pose an ethical
problem, because you do destroy potential normal human life. There's
no talking around it.
So if you accept that, then I would argue that a cloned embryo
is less of an ethical problem. That's all that I want to say.
So the potential is not there.
DR. HURLBUT: And from hearing you and talking with you
at lunch, I had the sense that you feel like nuclear transfer actually
offers us more versatility and more promise than using IVF embryos,
because we can use the genotypes we want and control various things.
Is this fair?
DR. JAENISCH: Yes, I would think so.
DR. HURLBUT: So I have to admit I'm troubled by your
comment, and not convinced by Michael Sandel's comment. As
much as I like my colleague Paul McHugh, I'm not—I'm troubled
by the notion that what you called not potentially normal human
life does not have a moral standing.
And the reason I'm troubled by that is—was just alluded
to by Gil, that I would not say that a potentially abnormal trajectory
of embryogenesis is not a human life necessarily.
Now, I have a handicapped child myself, and she does not have
a problem of development, but she had brain damage at birth. But
she's not what you would call a normal human life in that specifically
all-healthy sense. But she's very human, as human as I am.
I'll let somebody make the joke on that one, if you want.
But the question is this. I feel for what you're saying.
I feel for what Michael said, and I feel—have been with Paul to
some extent on this all along.
But if we say that natural—the natural meaning of the definition
of a human life is not normal, but some meaningful process of integrated—of integration of identity, of continuity, then it seems to me
that the clones that have been produced at least with animal studies
have to qualify as entities of their species.
So even though, as you said to me even a year ago, a cow standing
on a green hill chewing its cud isn't—doesn't make it
a normal cow. I agree. But I think it's still a cow.
And likewise, the human trajectory of whatever the clone could
be is troublingly human-like. I personally wouldn't feel comfortable
saying, for example, as was implied by John Gearhart in his talk,
that we could use it to harvest out later tissues for more difficult
Well, let me back that up. All of my colleagues, and my own reading
on this, confirms to me that going from embryonic stem cells to
useful cells, tissues, and organs, is a long journey, maybe short
for Parkinson's Disease, but longer for more complex tissues,
and very long for—what we really want is organs and—for transplant,
like kidneys, and so forth.
I think there will be a temptation in our civilization to move
the trajectory of something that's designated non-human on to
further stages of gestation. Now maybe not gestation in a human
womb. Maybe it'll be in an artificial endometrium or something
that can coax it just 30, 60 days longer for good use. It starts
to be troubling, even if you accept your premises.
Let me back it up, then. I think in principle what you're
saying has something to it, and Paul and I have been trying to talk
about this, and I've had dialogues with my colleagues on this.
I put forward in my personal statement on the President's
Council Report on Cloning what I called a speculative proposal,
where I suggested that—drawing on what Paul had said, that this
is an artifact, not a human being, a clonal artifact.
Why don't we go to the bottom of the problem? Why don't
we say that something that does not have a human potential in any
meaningful way, not an abnormal human potential, but no meaningful
human potential, might supply us with the essential way to get around
the moral problem, at least to get us to stem cells? And we can
deal with the next issues later.
And I want to stay two things before I conclude. One is in making
this proposal, I was very aware that the first issue of what you
might call the inviolability of the human life is only the beginning
of the questions about the moral use of human process or human tissues
But what I had in mind was this—that just as we've come
to see that a cell does not define the locus of human dignity, nor
does a gene, of course—let's start at the smallest. A gene—a human gene can be put into a bacterium, grown, used to make
insulin. A cell—we do blood cell transplants. Tissues—we do
skin transplants. Organs—we do transplants and even organ systems.
None of those are the locus of human dignity, human moral standing.
Likewise, in this age of developmental biology, we will learn
that partial generative potentials are not the locus of human moral
standing. We will harness these partial generative potentials in
such a way that we can do wonderful things with them without violating
the integrity, identity, and continuity of human life.
So this is my question to you. Could we find a scientific way
to meet the challenge of our President and our nation and the imperative
of our civilization to go forward with the science? Could we define
the danger as we never want to interrupt a normal human life in
process, or even one that is remotely normal?
Could we meet that moral objection while at the same time opening
the positive future of our science? And here the suggestion I would
have is using something so fundamental, like short, interfering
RNA, to preclude the very possibility of anything but the most minimal
and genetic mammalian process to the blastocysts from which we could
then take embryonic stem cells.
What do you think?
DR. JAENISCH: Yes. I think we talked about this before.
I think it's a really interesting possibility. So what we need
to—for a human being to develop two major lineages, which is a
trophoblast lineage and the epiblast lineage. The trophoblast lineage
will support the embryo from the placenta.
So certainly there are genes which are only needed, as far as
we know from experiments, that are only needed for placental development.
They are not needed for the epiblast lineage or for the embryo.
So one would modify a donor cell—let's say a skin cell that
you want to transplant—with inactivating such a gene—I can give
you examples of those—and expressing, for example, using the SI
RNA technology, which now really becomes—seems to become a routine
procedure to inactive genes.
And, indeed, an embryo—a cloned embryo derived from such a nucleus
would not be able to form a trophoblast lineage which would be functional.
So, therefore, if this embryo would be implanted, it would definitely
fail. You could predict it would fail much earlier than this continuum,
which I outlined for the cloned embryos which are not modified.
So I think, in principle, this is a doable—potentially doable
manipulation which would prevent what you are interested in and
what you are raising, prevent the development very far in utero.
DR. HURLBUT: In your opinion, could we reasonably call
that not in any sense a human life in process, abnormal or normal,
but call it a clonal artifact, a laboratory production for the procurement
of cells with a potential for something, but not what we would have
to call embryonic cells?
DR. JAENISCH: Yes, because you would argue this embryo—I mean, the definition of an early embryonic cell, like a two-cell
or a four-cell embryo, the blastomere, is totipotent. These cells
can make the placental lineage and the epiblast lineage.
This embryo would not be able to make the trophoblast lineage,
so, therefore, it would not be totipotent. So I think you could
argue this would be a real biological difference.
CHAIRMAN KASS: We do want to go to the next session shortly,
but I've got Janet and Elizabeth, and then I'm going to
call a halt for this session.
DR. ROWLEY: I come back to John Gearhart. And, first,
I want to take you to task for comments at the very end of your
talk, and then I have a question. And what I want to take you to
task for are your comments, all related to enhancement and how easy
it's going to be to just get genetic enhancement of all sorts
And I think—well, you could expand on that or explain what you
meant by that. But this has been such a critical issue for this
particular Council—the likelihood of enhancement—that I think
that it shouldn't just be left unchallenged.
And the question that I have is, with regard to characterization
of current lines, you indicated that that's under the auspices
of the MRC in London. And I'm just curious, are they looking
at both so-called presidential lines as well as lines available
elsewhere? Or are they only lines available elsewhere? And what
kinds of characterization are they proceeding with?
DR. GEARHART: Let me answer the non-controversial question
first. This is a relatively—this panel—there has been, over
the last nine months or so, an international consortium that has
been meeting on human embryonic stem cells.
The outcome of this was the appointment of a panel headed by Peter
Andrews at Sheffield University, who will be—that panel will be
setting up the standardizations, if you will, or the criteria for
the embryonic stem cell lines.
They will—I think some of this work will be done in the equivalent
of what's the FDA in the United Kingdom. But they will take
any cell line that's submitted to them, whether it's a presidential
one or one from anywhere else. They are not a—they are not distributing
cells. They are just testing them for these different parameters.
And, obviously, there will be a number of transcription factors,
a number of surface antigens. This kind of thing will be in that
DR. ROWLEY: Will they do karyotypes?
DR. GEARHART: Oh, yes, karyotype is—yes, it's absolutely
essential to have a normal karyotype, and one that's stable
over a number of cell passages is critical.
The comment I made about enhancements, I just—it may have been
out of order here, but I am concerned, I mean, as a biologist I'm
concerned. I'm concerned about a number of issues. It makes
it sound as if most of the time that scientists aren't concerned,
that we're just going along in a fashion that, you know, science
for science sake.
And I only have to remind you over the last few months to see
the use of human embryonic stem cells in mouse chimeras, the use
of chimeric human embryos that has been done, that you say there's
no reason that these things should have been done. I mean, there's
no scientific basis for this.
And we have to have some degree—and I'm not going to call
it regulation, I don't like that term, obviously, but there
has to be some consensus as to what should be permitted and shouldn't
be permitted in almost a global way. And I realize that this is
very, very difficult.
I've been to two G-6/G-7 meetings in which we have tried to
get some kind of consensus even on human cloning among countries.
I mean, let alone the use of embryos for research, and things like
this. And as you can imagine, nothing much happens. Everyone has
their own opinions, and we go away.
When we have the capability—if this work is successful—I'll
be frank with you. If this work is successful, and by success that
we are learning mechanisms around differentiation and the development
of cell types, we are going to have an awful power that we can instruct
Now, we're not necessarily going to be talking about germ
line cells or germ line modifications. Maybe. Because, obviously,
if we can grow oocytes from these cells—there are reports that
we can grow spermatogonia and sperm cells out of these—maybe the
possibility exists for that at this point.
But I'm just hoping that people are thinking ahead a bit,
that, you know, this is something that's going to be on the
table in a number of years, and that we should give it some thought—where we're going to go with this. That's all.
And so I made a comment about, you know, getting it off the internet
or something to do it. Maybe that's an overstatement. Maybe
that's just crying wolf, and I apologize for that.
But I think it's something that we have to be aware is out
there. It's going to be out there. Not out there now, but
will be out there, and we should be thinking ahead. So I apologize
to the Council for making that kind of a comment.
DR. ROWLEY: Well, I don't think an apology is necessary.
I think what I'm trying to say is—my own personal view is
that the Council should be especially concerned about things that
are relatively near term. And by "near term," in such
a rapidly-moving field I think a few years is near term. And if
you're saying, well, in 10 years' time we'll be able
to do this, we can't see that far ahead.
DR. GEARHART: Right. Right.
DR. ROWLEY: And so that—so what kind of terms are you
talking about in terms of time?
DR. GEARHART: Oh, I'm talking not about 10 years.
I'm talking about a shorter time interval.
DR. ROWLEY: Okay.
DR. GEARHART: You know, Janet, I think to be honest with
you, the number of federal Blue Ribbon Panels that have been set
up, at the NIH level, the National Academy, to discuss issues that
seem to be 10 years away, recommendations were made by our leading
scientists, our leading scholars, only to be ignored. And then,
we find ourselves in a pot once these things happen.
And it's just, I mean, frustrating as a scientist. It's
frustrating almost as a citizen to do this, where it seems that
the scientific community, to a certain degree, is marginalized when
it comes to making recommendations as to how we should proceed.
And we are left, then, every time something comes up in this area
everything is thrown up in the area as to where we're going,
etcetera. So that's the nature of the comment. Or, I'm
sorry, that's the basis of the comment that I made.
CHAIRMAN KASS: Last comment. Elizabeth, if you—did
you still want to say something? You were on the queue.
PROF. BLACKBURN: I'll return to the science, and I don't
know who would like to answer this question. But it was picking
up on the point that I think Leon did raise about asking about the
sources of adult stem cells. And you mentioned that there's
really only one good one, or the best one, is that right. Dr. Verfaillie's.
DR. PRENTICE: I think Dr. Verfaillie's is the best
PROF. BLACKBURN: Yes, best characterized.
DR. PRENTICE: —in terms of all of these various traits.
PROF. BLACKBURN: Yes. So, you know, we're struck—and I
think Leon pointed this out, too—by the fact that there's
a great deal of research that's been ongoing in the last, you
know, couple of years—
DR. PRENTICE: Yes.
PROF. BLACKBURN: —on adults. And yet it seems to me there's
a real problem. And, you know, what's the problem? Why is
there so little known? To put it bluntly.
You know, here's 192 publications that you cite, and why is
it so difficult? Are these just inherently really difficult things
to work with? Is this what it comes down to, that people who work
with adult stem cells don't really like to say that, or—sorry
to cut to the point, but I know time is of the essence.
DR. PRENTICE: I think one of the reasons comes back to
this difficulty in identifying which one is actually the stem cell,
how to actually purify it, isolate it. Some of the cells seem rather
easy to isolate. Certainly, peripheral blood is quite easier, or
the adipose-derived cells would be easy to isolate.
But then, how do you really know that one is the stem cell? And,
again, most of the results—the best results seem to come primarily
from these mixed populations of the bone marrow stem cells, where
they have not intentionally in many cases isolated the particular
stem cell. Instead, they've been geared towards some sort of
My assumption would be, then, working back from that, once they
could achieve that end point, then they might be able to derive
the particular cell.
But I think it also brings up a point I tried to raise before
in that it may not be possible, or it may not be preferable, to
try to get a particular cell. It may be the context of these cells
with the population and their interactions, exchanging growth factors,
altering their own gene expression, and then—in the tissue milieu—that really gives them this ability to somehow cause the repair.
And I use that term because, again, I think in many of the instances
the research is showing that it's not direct integration and
taking on the function. It's somehow stimulating endogenous
PROF. BLACKBURN: So it may—are you hinting that it might be
qualitatively different from what is going on with embryonic stem
cells, when those were tried to be used?
DR. PRENTICE: Well, I—
PROF. BLACKBURN: I don't want to get too much into this,
but I'm just trying to understand the science here.
DR. PRENTICE: Yes, I think that might be a good way to
put it. It may be qualitatively different. You know, it's
interesting Dr. Jaenisch mentioned that embryonic stem cells, in
a sense, are a tissue culture artifact. He might just as well,
by the same definition, call Dr. Verfaillie's MAPC cells a tissue
And it may be that putting these cells into culture, now changing
their context, gives them some of these characteristics that are
so sought after. Our target might be better put in terms of, what
is the physiological endpoint that we want? Not a particular starting
cell, but how do we achieve the repair of the tissue damage?
And, again, as I mentioned in the talk and in the paper, it may
be that it's not a particular cell that we really need. Short
term, cellular regeneration may be our best goal. But long term,
figuring out the particular molecular and cellular signals—and,
again, it may be a context-type signal—cell-cell context or cell-matrix
context—that actually can give us that physiological endpoint,
and so eventually going to a non-cell-based system, but a signal-based
system if you will.
CHAIRMAN KASS: Paul McHugh insists on a very brief comment,
and then we're—
DR. McHUGH: I have just two—
CHAIRMAN KASS: —going to break.
DR. McHUGH: —little tiny scientific questions, but because
they are important science that we're discussed here. The first
one really relates to Dr. Jaenisch, and that is, I wanted to know,
sir, now whether you could say that the genetic alterations and
the problematic genetic status of the fully cloned animal, whether
those genetic abnormalities would then pass on to their offspring,
and so, therefore, be a contaminant to the whole human genome legacy
from then on? That's the one question.
DR. JAENISCH: Let me first clarify there are no genetic
alterations which are important. They are all epigenetic.
DR. McHUGH: Right, yes.
DR. JAENISCH: So the experiment has been done. You can
take two animals afflicted with large offspring syndrome and mate
them. The offspring will be all (risk?) normal, because you send
the genome through the process of gametogenesis, which reprograms
everything in a normal way, everything is reset.
Of course, there might be some genetic alterations which are acquired
during somatic life.
DR. McHUGH: Right.
DR. JAENISCH: Those would not be—they cannot explain
the phenotype. They would be recessive, and so they haven't
shown up yet.
DR. McHUGH: Right.
DR. JAENISCH: So we don't know if they would show
DR. McHUGH: They might show up later.
DR. JAENISCH: Right.
DR. McHUGH: And secondly, to John Gearhart—John, you
whine a lot about how troublesome it is to be a scientist nowadays.
I just want to know this. The President made a very good Solomonic
decision and produced the cell lines for you scientists to work,
putting the ball in your court.
Are you saying to us now you've used up all of the potential
in those cell lines, and now, because of the President's decision,
are restricted in your progress?
CHAIRMAN KASS: You don't have to answer, and he's
over time. But if you'd like to—
DR. GEARHART: I have to answer Paul.
CHAIRMAN KASS: —15 seconds.
DR. GEARHART: Being a scientist is the greatest thing
in the world, and I wouldn't trade it for anything, Paul. And
so I don't—
DR. McHUGH: I know that, John. Me, too.
DR. GEARHART: Okay. And I'm not here to whine. The
funding issues are improving, and they are improving nicely here.
Okay? And, clearly, we can learn a lot through the lines that are
CHAIRMAN KASS: Gentlemen, thank you very much.
Look, we have a guest we've kept waiting for quite a long
time. Would the Council, please, 10 minutes.
Thank you all very much.
(Whereupon, the proceedings in the foregoing matter went off
the record at 4:10 p.m. and went back on the record at 4:23 p.m.)