38th Annual Killian Award Lecture—Rudolph Jaenisch

Search transcript...

MATTHEWS: I am June Matthews. I'm a professor in the Physics Department and serving this year as associate chair of the faculty. And I have the distinct honor of introducing this year's recipient, the 2009-2010 Faculty Achievement Award winner, Dr. Rudolf Jaenisch, professor of biology and Whitehead Institute founding member.

Professor Jaenisch is a pioneer in the field of mammalian developmental genetics. He has made landmark contributions to his field year after year, decade after decade, throughout his 40-year career. Indeed, as the letter writers for his nomination case note, Jaenisch is remarkable for his, quote, "perpetual youth and his tastes and instincts, time and again, to conduct research in cutting-edge areas."

Very early in his career, Professor Jaenisch helped found the area of transgenic science, the science of gene transfer for making mouse models, or transgenic mice, now widely used for studying human genetic diseases, including cancer and neurological disorders. This work became the foundation for his subsequent research and discoveries in stem cell biology, mammalian cloning, and the epigenetic regulation of gene expression-- work which has opened up new horizons in stem cell therapy and regenerative medicine. Professor Jaenisch's most recent breakthrough is in cellular reprogramming, and he'll tell us something about that a little bit later.

He has developed strategies for reprogramming fully differentiated adult cells into unspecialized stem cells called, "induced pluripotent cells," or "iPS cells," which have the capability to grow into any cell type in the body. This work has enormous potential for the study and possible treatment of human diseases through the possibility of growing healthy cells from a patient's own cells.

Professor Jaenisch has received awards almost too numerous to mention, but I'll do so anyway-- including the Boehringer Mannheim Molecular Bioanalytics Prize in 1997, the first Gruber Prize in 2001, the Koch Prize in 2002, the Brupbacher Foundation Cancer Research Prize in 2003-- I think we're getting into, sort of, a prize per year here-- the Max Delbruck Medal in 2006, the Vilcek Prize in 2007, and the Massry Prize in 2008. He is a member of the National Academy of Sciences, a fellow of the American Academy of Arts and Sciences, a member of the Institute of Medicine, a member of the German Academy of Natural Sciences Leopoldina, and an associate member of the European Molecular Biology Organization.

Professor Jaenisch's passion and commitment to his work extends far beyond his lab. He has taken a leadership position in the controversies and debates surrounding cloning, and helped educate the public on the important distinctions between therapeutic cloning, which involves the use of stem cells for curing disease, and reproductive cloning. He has participated in numerous panel discussions, held interviews with the media, testified before Congress to help provide a scientific and ethical basis for government decision-making on cloning.

At MIT, Professor Jaenisch is a committed citizen, a popular teacher of both undergraduates and graduates, and a caring mentor. His students and colleagues are inspired by his creative ideas and thinking, and grateful for his wisdom and advice. As a founding member of the Whitehead Institute at MIT, he played a key role in shaping the Whitehead into the world-renowned biomedical research center that it is today.

Professor Jaenisch's contribution to science and society are a great source of pride for MIT. He is eminently worthy of the Institute's highest Faculty Achievement Award, the 2009-2010 James R. Killian, Jr. Faculty Achievement Award. So will you please come forward, Professor Jaenisch?


I would like to present you with the actual award, signed by the president and Thomas Kochan, chair of the faculty, who could not be here today because he's teaching a class. So we have our priorities right there. "James R. Killian, Jr. Faculty Achievement Award for the academic year 2009-2010 to Rudolf Jaenisch in recognition of his landmark research in mammalian developmental genetics and transgenic science, and for inspiring, teaching, and mentoring generations of MIT students."

JAENISCH: Thank you very much.


MATTHEWS: Now, he will give us his eagerly awaited lecture. The title on the poster says, "Stem Cells, Reprogramming, and Personalized Medicine, Promise, Problems, and Reality."

JAENISCH: OK. I got the wrong title. Really, thank you very much for this generous introduction, and thank you for coming. I must say, it's a really great honor for me to be chosen for this award-- to be chosen by your colleagues, whom you can't deceive. They know you, right? It means much more than any outside honors, so I am really, very honored. But it also puts pressure on you. I think it's much more pressure to talk in front of your home institution than anywhere else.

So I thought before I get started with the topic of my talk, I want to spend a few minutes to point out some of the key people who shaped my thinking and were certainly important for my career. I studied medicine in Munich, and I must say, I really hated it. It was overcrowded. It was memorizing. No challenge. So my escape from this was trying to get in the laboratory and do some experimental research. And it was a long, long way for me to become, again, interested in medical problems, as I am now.

I started in the laboratory of Peter-Hans Hofschneider, who had one of the few laboratories in Germany at this point that worked with molecular biology. And I did my thesis on replication of small DNA phages [? phiX174. ?] When I finished my thesis and got my MD degree-- I finally got it-- I became a postdoc at Princeton, in Arnie Levine's laboratory.

Arnie just had set up his own lab, working on SV40, a tumor virus, and I was his first postdoc. So when I came, after two weeks, he told me, well, by the way, I am going on a sabbatical to Europe. You run the lab. I was rather shocked.

So we worked on SV40 DNA replication, which has the same size as a bacteriophage. So it was really, for me, rather familiar territory, to learn something about replication. What I was really interested and puzzled by about this tumor virus was, why did this virus make only sarcomas, when you put it into a mouse, and not brain or liver tumors? So the old question of tropism of a virus.

And so, I thought, well, maybe the virus cannot infect liver cells and brain cells. That's why it cannot make tumors, or does infect them but cannot transform them. So I puzzled about this. And then, I read a paper which really was shaping my whole thinking.

The paper was by a prominent developmental geneticist, Beatrice Mintz, in Philadelphia. And she made a really incredible experiment [? of mine. ?] She used embryos from a black and white mouse, and aggregated those, and made what's called a "chimeric mouse." It's composed of all these cells. She did this to learn about the development of pigment formation, which I did not understand, as a simple molecular biologist at this point.

But it's a great paper. I use it every year in class, and, actually, we'll discuss it tomorrow in class. But I was fascinated by a totally different aspect. I thought, well, if you could put the virus DNA in these early cells-- the [? "blastomas," ?] as they're called-- they would end up in the liver and the brain, and I could answer my question.

So I got all excited and called her up the next day. And she agreed to see me. She was very friendly but extremely skeptical about-- [? it ?] [? wasn't ?] [? done ?] [? yet ?]. Who was I? So I was disappointed. Then, she called me back after a week and said, well, she thought about it. I should do it in her experiment.

So when Arnie came back from Europe, I told him this, and he said, you're nuts. But if you want to do it, you can. I will have no part in this. So you can do the DNA. So I just commuted between Princeton and Philadelphia, and Mintz showed me how to isolate embryos and how to use mouse genetics. It was, for me, the most important month, probably, in my early career.

So I learned it, and I got mice. And the mice were disappointingly, totally normal-- four legs and two ears, no tumors. So the question was, was it a real experiment, or was it a failed experiment? Did these animals carry the SV40 DNA, the tumor virus DNA? And this was not trivial at this point. There was no PCR. There was no Southern blotting invented. You couldn't even buy labeled triphosphates. So that was a problem.

I got, then, my first job at the Salk Institute. At this point, Paul Berg and Peter Rigby had invented Nick translation, which is a way to make very hot probes, and they showed me this procedure. And Tony Hunter, who was a Salk faculty, showed me how to make triphosphates in the lab. I had no idea how to do this.

So then, I did a Cot curve. Now, only very few of you might know what a Cot curve is. It's not used anymore. You melt your probe DNA and you let it reassociate. It takes days. And if you know [? it, ?] [? you ?] test DNA. If the test DNA carries also viral sequences, the kinetic is accelerated. That's the only way of doing this.

And I did this after killing the mice. I found, indeed, that these animals carried sequences in their various organs. So this led them to the really first transgenic mouse, although the name would be coined only six years later.

I was very excited. It was interesting, but there was a problem. The problem was, these mice didn't get tumors, although they had the DNA. And I didn't know, at this point, about epigenetic silencing of these viruses, which I'll come back later to. And I didn't get germ line transmission, because, in hindsight, I set up the experiment not correctly.

At this point, two postdocs from MIT joined the SALK faculty-- Inder Verma and Hung Fan, both postdocs in David Baltimore's lab. And they brought with them the Moloney leukemia virus system, which efficiently uses leukemia in mice. So I thought, this is maybe-- I should use that system.

And they showed me how to use it, and I did the experiment's visual outline in my favorite slide from this time. It's a cartoon made by Jamie Simon, who was a technician in Hung Fan's laboratory. What it says is exactly the whole experiment.

You take two mice, mate them, isolate embryos at different stages, expose them to a virus, and you have to now put them in to experience a foster mother to carry them to birth. Then, you want to know whether, does this have an integration of the germ line. You mate them. And then, you take the offspring, and you have to isolate DNA.

So the DNA isolation, we did by partial hepatectomy, and got that mice recover if you treat them well. And, by then, Southern blotting was invented. You could visualize the bands on the gel, or you could package them into a phage, infect an E. coli-- and they didn't know that's how [? scratchy ?] that is. And you can isolate the locus where the virus is integrated.

So that was the experiment, and so we did it. And, indeed, it worked. So I got mice, which, indeed, transmitted these sequences as a [? Mendelian ?] trait. So that was very exciting, and this experiment really shaped a lot of my thinking. Let me give you one aspect of this.

When the virus integrates, it can disrupt the gene, depending where it ends up. So it makes a mutation. At the same time, it marks the gene, because it can isolate the gene. It's called, "insertional mutagenesis." And this was a way, before genomics, to isolate gene-specific development. If you've got one of those-- which was a lethal mutation by insertion, by activating the collagen gene. [? Most strain ?] which we learned a lot about viral mutagenesis, and actually was done by Angelika Schnieke at this point.

So I used this a lot for coming years to study developmental issues, and I had moved to Germany by this time. And I should mention, really, one very important mentor for me. This was David Baltimore. David was interested in retroviruses.

He had communicated my paper to PNAS. And when he founded the Whitehead Institute, he offered me a position. And I was clearly to accept, and I came to Whitehead today, 26 years ago.

So my interest was, really, using viruses for development and studying development. Let me give you one other aspect here, which was puzzling. When you infect these early embryos with a virus, they cannot replicate the virus. They're totally unable to do this. But if you infect them five days later, they were really very efficient in replicating the virus.

So the conclusion was, it was a developmental regulation of virus replication. The early embryos were non-permissive. The later ones were highly permissive. And it was a puzzle. What was the mechanism? And I found out only later-- today, we describe it as epigenetic regulation or silencing. So what is it?

If you know the sequence of one cell of an organism-- of a human, you know the sequence of all cells. But this doesn't help you to understand how the genes work, because genes are packaged into chromatin if we have as many epigenomes as we have cell types. The key issue of the field is, what distinguishes one cell from the other, and how can you convert one cell to the other? Set the clock back-- reprogram [INAUDIBLE]

So that was an interesting-- there was a lot of progress now. But how does epigenetics work? Because there are non-biologists in this audience, I want to give you a very simple analogy, which comes from a text.

This paragraph-- "to be or not to be-- that is the question--" you can easily read, and you know it anyway, because this text is fully formatted. I want to make an experiment. I want to remove all punctuation, all spaces. Now, the information content has not changed, but it's really much more difficult to read.

That's exactly what epigenetics does to the genes. It make them readable or non-readable. In the early embryo, there might be some genes, for embryogenesis, which are readable, but those others not in the somatic cell, it's the other way around. That's what epigenetics does.

So what's the control? What's the mechanism of this? And it was known for a number of years-- the modification of one of the bases [? as ?] hydrazine, with this group, made methylcytosine. And it was correlated for a long time with gene inactivity, but this was all a correlation. And the question was, really, is there a causal role for this? And does it play a role in development?

I was at MIT then, and had the incredible luck to attract a brilliant student, En Li, to my lab. En made a number of knockouts. He made the first knockout. Among those, also, was knocking out the methyltransferase-- it's the key gene to setting up methylation. And the phenotype of these mice was very informative. They died very early-- meaning, methylation's essential for development.

This mutation was important for us, because it allowed us to study epigenetics by genetic means. We could study the role of epigenetics in cancer, in imprinting, and in reprogramming. And that brings me to the focus of my talk-- to embryonic stem cells.

Embryonic stem cells-- they are really important and interesting because they allow you [? tissue ?] engineering, they allow you to make any cell type of the body in the culture dish, which, of course, has really created enormous interest for medical application. They're generated from an explanted embryo-- from these cells, which make the embryo. You put them in the dish, and, in the right conditions, they might proliferate. And you call these cells, "embryonic stem cells."

And I want to show you in a movie, which I hope works. Yes. How this works-- these are human blastocysts, and these are the cells you want. The other cells are support cells which have to get digested away. So you take these-- what's called, the "inner cell mass cells"-- and they're put, now, in a Petri dish. And, as I said, in the right conditions, you might induce them to proliferate, and you call them, then, "embryonic stem cells."

And then, the [? key ?] is-- under certain experimental conditions for certain factors-- cytokines or so-- you can induce these cells to differentiate to different cell types, including liver cells, pancreatic cells, muscle cells, nerve cells-- any of these types of cells. And that's, of course, very exciting. Because, in principle, that provides you with a source of cells which you could use for treatment if you can solve the immunological problem. So people, of course, fantasized, this could be useful for treatment of a number of diseases, and many of those are probably unrealistic.

What is the goal of stem cell research? It would be to use the potential of these cells, to provide customized cells for customized tissue repair-- and maybe even for rejuvenation or for prolongation of life. So the induction of pluripotency and the quest for eternal youth is really not that new.

It was actually-- this is a painting from a German painter-- 500 years ago-- Lucas Cranach, who envisages the fountain of youth. We go on one side and there's an old, sick person. You rise on the other side, really rejuvenated. And you've surely thought about a sea of stem cells which would do that to you.

And, actually, this is something which, of course, has gotten the attention of the newspapers. And some really have seen remarkable treatment results. So, actually, I assume some of you might be in need of stem cell therapy, so I want to give you some [INAUDIBLE] proven [INAUDIBLE] that you can go to, which you get from the internet.

This is MEDRA. This very [? confidence-giving ?] doctor has treated, as he says, thousands of the most devastating diseases, including epilepsy-- that's amazing-- spinal cord injury, even Down's syndrome. Very impressive record. This stem cell therapy company, as a measure of success, quotes this Mr. Fernandez, who is 57 years old, as old as Bill Clinton at this point, but, due to stem cell therapy-- look how much younger he is. So whoever has a problem with graying hair-- this is the source you want to go to.

And then, finally, my final example would be the Swiss company. They treat everything, from burnout syndrome to impotence, to anything. And they're very imaginative. They use a Swiss Apple Stem Cell Serum, which is based upon a rare Swiss apple, an endangered species, the Uttwiler Spatlauber.

So this is really-- I cannot highly recommend this, but you'd pay for this. You'd pay 10,000 francs for three days of serum. And this doesn't include the view of the lake and food. Actually, 10,000 francs is pretty cheap. Many of these companies ask for $50,000 to begin with, and you're lucky if you just lose the money and not more. So it's really a concern-- this stem cell tourism-- for the field.

The concept of personalized medicine, then, is, use your own cells in transplantation medicine. And a major turning point was when Dolly, the first cloned animal, was generated in the late '70s in Wilmut's laboratory, and I was very pleased to see that the second author was Angelika Schnieke, who used to be in my lab before. I knew this was a very solid paper.

And then, a year later, Teru Wakayama, in Yanagimachi's laboratory, produced the first cloned mice, and I was really excited. I really needed [INAUDIBLE] and arranged for a collaboration with them, with Teru. And we introduced [? nuclear ?] transfer to the lab to study epigenetics.

And I was really, again, very lucky to have some terrific students-- Kevin Eggan, Konrad Hochedlinger, David Humpheries, and a postdoc, [? Bill ?] Rideout, who really were driving the nuclear transfer in my laboratory to understand how the egg [? manages ?] this. So what's behind this technology?

You have to remove, first, the egg nucleus, and replace it with a nucleus from from the somatic donor cell. Then, you incubate these to make a cloned blastocyst, and, if you put it in the uterus of an animal, you get reproductive cloning. You get Dolly. If you put it in a Petri dish, as I said, you get, really, patient-specific or donor-specific embryonic stem cells.

And I want to show you how that works. There's a movie of Kevin. He was really a master in this. Shows you, in mouse eggs, how you do this. This is the enucleation step. You suck out the metaphase plate-- all the DNA in that. And, as you can see, it's not that easy to see where the-- he has to turn it around so that the nucleus faces his pipette. And you [? really ?] can't see it very well there. But he goes in there, through the egg [? chamber. ?]

Then, the next step would be to generate the donor's nuclei, where you go with a smaller pipette to suck up the cell, disrupt the cytoplasmic membrane, and take up the nucleus into the pipette. And then, the final one would be to introduce the nucleus into the enucleated egg. So, again, you have to go through the egg shell and deep into the cytoplasm to deposit one nucleus, because you want to get a diploid clone.

It looks pretty simple. It's not a simple technique. It's very high-power. So there the nucleus comes, and you won't have any nuclei-- you want to deposit one nucleus into this nucleated egg to make a clone. Then, you incubate these. They cleave. They form a blastocyst, which now has to be implanted. And out comes, then, cloned cows, mice, cats, pigs, and Dolly.

About 20 mammalian species have been cloned by now. Now, this experiment created an intense public debate. And the debate was, should we use this for human application? There were many reasons, given that it was a very public debate. Many people have very clear opinions on this, and I'll give you some of the reasons.

Maybe the most serious one was, it would give childless couples the opportunity to have a genetically related child-- at least, to one parent. Then, some thought, well, it's a way to wake up dead people. But it gets worse. Some people thought, well, it's like a repair kit. You have a clone, you need a heart, you cut out the heart, right?

But then, of course, people thought about eternal life, and even dealing with terrorists. I want to give you some examples of these arguments, because they are rather interesting. This is Rael, the leader of the Raelians, a sect in Canada.

They think life came from outer space, by cloning. And he suggested in public-- and that's why many people go to his sect-- that they promised to make exact genetic copies of yourself, then, sort of, download your memories into the clone. And this way, you'll come to eternal life. Many people got attracted by this.

But he suggested something more interesting, I think-- that this will really deal with terrorists. And the idea was the following. It would allow us to clone terrorists before they do the act. And you have several. And then, you could use one clone for waterboarding, one for hanging, one-- so they would not escape. And he suggested this, actually, three days after September 11.

Actually, it was interesting that these people testified in the United States Congress. So there's Zavos, one of these characters from Kentucky. This is Rael in a spacesuit, telling representatives about eternal life. This is a doctor from Italy. This is Boisselier, the chair of Clonaid, which, together with the Raelians, offer you, for $50,000, to clone you. And they testified, and I happened to be at this hearing. And it was really, like, a happening. And it was quite interesting.

Anyway, what I learned in textbooks about development is, it's unidirectional. What Dolly told us was, you can reverse this. What, really, that said is that the nucleus of a differentiated cell maintains the potential to generate a whole organism.

Now, if you put nuclear transfer and embryonic stem cells together, you come to the concept of therapeutic cloning. Putting this together, of course, initiated enormous debate. So let me just tell you how that works.

This would be an embryonic stem cell made from a fertilized embryo. You can make therapeutic tissues of this. But if you would use for therapy, it will be invariably rejected, because it's genetically different from the patient.

But if you generate a cloned embryo, cloned from the nucleus of the patient-- so, an asexually produced embryo, you make a customized therapeutic tissue. If you inject those cells into the patient, they will not be rejected, because it's a patient's own cells. So this was the concept. And there were many promises made-- big promises-- and the question was, could they ever be useful? Could they be ever realized?

In mice, we checked that. And we made a proof of principle experiment, when George Daley was still at the Whitehead, using as a patient a mouse which has a genetic defect-- has no immune cells. It has a severe combined immune deficiency. There were two steps to the experiment-- nuclear transfer and ES cell derivation.

So you would make a cloned blastocyst, and then an embryonic stem cell. Then, you would correct the gene defect by simple targeting. The second step would differentiate the cells to hematopoietic stem cells, and put them back into the mouse, and then restore the immune system. So the experiment worked. It was exciting.

But, of course, when you replace the mouse with a human, this is not an option. It doesn't work in humans, and major ethical problems-- how do you get human eggs? It's simply, totally impractical. So the only way was to learn, how does the egg accomplish this? Can you do it without the egg? So can you do that directly?

Many laboratories were really, highly interested, including my laboratory. And then, four years ago, you might remember, Shinya Yamanaka published his landmark paper, where he generated iPS-- induced pluripotent cells-- without the help of the egg. I want to tell you briefly what this involved.

It was based upon the knowledge, what drives self-renewal of embryonic stem cells, and what drives differentiation? And numerous labs have contributed to this. Among those, prominently, was Rick Young's laboratory at the Whitehead Institute.

And this regulatory circuitry really says there are three transcription factors, very much upstream of this whole circuitry. And those-- key-- are called, Oct4, Nanog, and Sox2. And this knowledge was used by Yamanaka.

He took fibroblasts, which carried a genetically engineered, endogenous Oct4, one of these three genes. It was fused to anneal mice in drug-resistant gene. This Oct4 gene is off in any somatic cell. So when you put the drug on these cells, they will die.

So then, what Yamanaka did-- he transduced four genes, by retroviral transfer, into these cells-- Oct4, Sox2-- two of these genes which I just mentioned, and two oncogenes-- Myc and Klf4-- and selected for activation of the endogenous Oct4 gene by selecting for drug resistance. So this gene turns on. Then, the cells become drug-resistant.

In this way, he got what he called, "iPS cells." Now, these original cells, in 2006, were really not ES cells. They were partially reprogrammed. So people were really skeptical. How could it be? People just didn't believe it.

But then, a year later, three groups independently showed that iPS cells were identical to ES cells. And this was the group of Yamanaka, my group, and the group of Konrad Hochedlinger, here at MGH. Three groups independently doing this-- people had to believe it, and it really electrified the field.

And then, really, a few months later, human iPS cells were isolated. This really changed the field. The cloning concerns went away, and nobody really talks about Raelians anymore. They became irrelevant.

I was lucky. Again, there were a number of really excellent people in my lab who drove this. It was Marius Wernig, Alex Meissner, Ruth Foreman, Tobi Brambrink, who spearheaded these efforts. They've [? all ?] [? left ?] now.

And this would be a typical picture of an embryonic stem cell-- these small, very tight colonies-- and the iPS cells look very similar. And they, indeed, resemble-- by all molecular aspects and morphology-- the ES cells. But really key was, are they useful for therapy? Could they function-- Could they differentiate to functional cells? And one simple way to do it is to differentiate them to heart muscle cells, which then [? contract ?] in the dish.

So that was fine. That was reassuring, but it was not enough. If you want to know whether these cells are pluripotent, the key criterion was, could they make mice? You make mice by injecting these iPS cells into a host blastocyst, into this cavity-- about 10 cells.

And they get in and they incorporate themselves into the embryo, and then participate in the embryo. And then, you get out what's called, "chimeric mice." The [? group ?] brown color is from the iPS cells, and the black color from the host.

So then, you have to ask the question, of course, can they also [? contribute ?] to the germ line? So can it be transmitted? So you mate these, and there you are. You get the brown [? pups ?] as a dominant coat color, showing that, indeed, these cells could be transmitted to the germ line.

If I summarize this approach-- these are cells growing in a culture dish. You take a single cell, introduce these four factors, and you get a whole organism out of this. A mouse with ears, eyes, and four legs. It's quite amazing.

This is really an amazing experiment which, I think, has changed a lot of how we think and how we approach biological problems. And obviously, it has enormous implications for medical application.

And so I want to go switch to that part now. So the idea would be-- in a clinical setting of a patient suffering from some disease, you take a skin biopsy, generate these cells, and now, depending on what the disease is, you would differentiate this to the cell which is defective in these patients and hope you can learn something about the mechanism of the disease. Because these cells behave differently in the culture dish.

If you find this, you might find a drug which helps, or, eventually, even to generate cells which could be used for therapy. But there are a number of issues which need to be resolved to make this, really, a viable approach. And I want to just point out three.

First, we need to get rid of these vectors. Secondly, gene targeting is a major problem, as I tell you. And finally, you want to know, what is the best cell for using as a donor? So let me just briefly address these three questions.

Genetically unmodified cells-- no vectors, of course-- key. These viruses can make mutations, as I told you. They can make cancer. So people have used different strategies. One is to use vectors which do not integrate or can be excised. It works quite well.

Better would be to not use any DNA. So use of proteins, which people have done-- modified proteins which can get into the cell. There's no genetic modification. It is very inefficient, but it's a really interesting approach. Or small molecules-- which also is quite successful already in replacing a number of these factors. So I think this is pretty much solved. It has to be more efficient.

Gene targeting is absolutely routine in the mouse. It's almost a trivial experiment to do in drives, but it's very difficult in human ES cells and iPS cells, which really impeded progress. So we thought a different approach is needed, and we used zinc finger nucleus approach, which I just want to briefly outline.

Zinc fingers are transcription factors. They bind to specific sequences, or, in this case, they will fuse to a nuclease. And when this pair binds close to each other, the nuclease is dimerized, and this introduces the double strand break, which makes a mutation if you don't do anything else. But if you offer a vector which is homology to the ends, this can integrate at the site of integration.

So it's a different means to genetically manipulate cells, and this was done in collaboration with Sangamo BioSciences. And in my laboratory, it was spearheaded by Dirk Hockemeyer and Frank Soldner. And I just want to summarize what the approach did.

They used, in the initial experiment, three genes-- Oct4, the one I mentioned. It was almost 100% targeting efficiency. That was pretty good. This locus is expressed in ES cells. This is a very useful locus for expressing predictably a transgene-- 50% percent targeting efficiency. And Pitx3 transcription factor was not expressed in ES cells, but gets activated in dopaminergic lineage, which is the lineage in the [? neurons ?] which die in Parkinson's-- 10%.

That was pretty reasonable. This approach is efficient. You need very short arms of homology, and no need for vectors which come from the cell you want to target, which you need for homologous recombination. So we believe this is a good approach to help us. And we've used it now for quite a number of other genes.

The third issue was, in a clinical setting, what would be the most suitable cell to make iPS cells? Now, at the moment, the cells come from punch biopsies of your skin. It'd be much nicer if you could do it with peripheral blood. That's the most accessible setting in a clinic.

So we tried that. And I'm not gonna go into the protocol. It takes about four to six weeks until you get these cells, and they have all the right markers, and the experiment [INAUDIBLE] And so, let me just summarize what the usefulness of these blood-deprived iPS cells is. They would be, of course, much easier to use in a routine generation of iPS cells. But more importantly, in tissue banks, there are large collections of blood from patients.

Now, with these bloods, you can't do anything to study the disease. But if you can make iPS cells, you could. So that is of interest to us. If you think about the technical challenge of this technology, it was gene targeting, elimination of vectors, and donor cells for routine clinical use. I think those might be resolved.

The fourth one is a key one. Namely, can we make this useful for studying the disease? Can we see a phenotype which is relevant in the culture dish? If this is a patient that has some disease, you make an iPS cell.

Then, depending on what lineage you're interested in-- in Parkinson's, you might be interested in the neural lineage , in a skin disease, in the skin lineage-- you want to make those and differentiate along this lineage and see whether they teach you something about the disease. Is something different as compared to a cell which comes from a normal individual? And so, I'll give you an example.

ALS is a motor-- motor neurons die. It's a terrible disease. You could make, and people have made, ALS-specific cells, and compare those with ES cells or iPS cells from a normal individual. Then, you can differentiate those cells in the culture dish to motor neurons. And the question is, would they behave differently when they come from a patient than from a normal individual? Would they, for example, die earlier or have some other normalities?

And, actually, they do in some model experiments in this disease. So that's very interesting. Because, then, you can ask the question, are the changes in these cells similar to those which occur in the patient? If you do find that, then you really, I think, have a great opportunity to screen chemical libraries to find a drug which slows or prevents the degeneration in vitro, which will be a direct path to a therapeutical to [? the protocol. ?]

So we were interested in human stem cells, and we established a number of human ES cell [? lab ?] at the Whitehead with, of course, independent funding from federal funds. And it was really made by Maya Mitalipova, who was very experienced in isolating human embryonic stem cells from embryos. So she did this and taught many people in the laboratory, including Frank Soldner, how to use those.

We were interested in Parkinson. Let me just give you an example of some results from the Parkinson. We made a number of Parkinson iPS cells from various aged patients. They were all nice. They had all the properties we expect. They were done by Frank Soldner again to [INAUDIBLE] and the question was now, can you differentiate these cells to dopaminergic neurons? Would you see a phenotype? Because these are the cells which die in the patient.

So we did that, and we used several protocols to do this. And I don't want to go into any details, because the answer is very clear and unambiguous. There was no difference. Now this might feel disappointing. But I think it's rather an expected result, because this is the problem of the whole field.

A disease like Parkinson develops over decades. [INAUDIBLE] iPS cell-- we don't have decades. We have weeks, or maybe a month or two. So the key issue of the field is, I think, to learn how to differentiate the cells of the right developmental progenitors, [? or ?] the embryo [? users-- ?] adult progenitors, but maybe more importantly, to expose the cells to stress situations, which people believe play a role in the disease, such as exposure to pesticides or exposure to oxidative stress, and, this way, try to telescope the aging process, and maybe see something which is different in the patient-derived cells from the healthy individual.

This would allow us to study the disease. So where are the issues? What tools do we have to develop? So as I said, at present [? views ?] viral transduction, which is not acceptable. So, clearly, we have to do it for the [? old ?] viruses, and I think we're on the way to doing this. But very important will be genetic manipulations.

We have to insert markers like GFP into developmental regulators to teach us how to differentiate cells to a given [INAUDIBLE] So the zinc fingers might help us to do this. Cells in the body don't grow on a two-dimensional dish. And three dimensions of tissue engineering will be very important to possibly get the cells to phenotype.

But this all may not help us. We may not see a phenotype. What do we do then? I think the next step will be important-- to make human-animal chimeras. The idea would be to take iPS cell-derived neural precursors, for example, from a Parkinson patient, and induce them into the developing brain of a mouse, and make a brain which consists also of neurons derived from the patient. And then, you can study over the lifetime of the animal the properties of the cells.

I am aware this is an experiment which has hurdles-- approval hurdles, and so on and so forth. But I think it's absolutely essential. And we're doing those experiments now, but it's clearly, technically challenging.

So this was what we do in the test tube. How about-- can we think about cell therapy? And I want to just briefly end with two experiments which prove a concept for two diseases-- sickle cell anemia and Parkinson's.

Sickle cell anemia is caused by mutation of a globin gene and makes red blood cells-- sickle and lice. And they clog up the small vessels, and these patients have brain and splenic infarcts-- very severe anemia. It's really a very, very severe disease.

Tim Townes from Alabama made a humanized model of sickle cell anemia by removing the mouse globin genes and replacing them with this human-- either the sickle version or the [? whiter ?] version. And then, when he made them homozygous for the sickle version, you can see the proof of that-- these red blood cells, sickle, and they have these huge spleen infarcts. This is a model which truly resembles the human disease remarkably well.

The experiment was, then, to take such a sick-looking mouse-- this anemic, really miserable-looking mouse, take a biopsy, transduce [? him ?] with these four factors, and generate iPS cells. We were worried about these oncogenes, Myc, because Myc makes leukemia. We were worried about giving these mice leukemia, so we had engineered the Myc so we could excise it after [? doing ?] this [? job. ?]

And then, we got these Myc-free iPS cells, and we had to correct the mutation by gene targeting, and then differentiate these cells to bone marrow stem cells and put them back into the mouse. And you can see this remarkable recovery. That looks like a happy mouse. And this experiment was done by a great postdoc, Jacob Hanna.

Let me give you what a resident would do if you were admitted to a hospital, and test in comparing all control mice-- heterozygous, [? to ?] the experimental ones-- you can see they are highly anemic. The hemoglobin is very low-- 50%. The red blood count is less than 50%. The hematocrit is reduced. The immature blood cells are very highly increased, the bone marrows are constant stimulation. The animals don't gain weight.

But, really, either eight or 12 weeks after transplantation, you see this is largely restored to normal values-- 80% to 90%. The red blood count, for example, goes up to 80%, 90%. Hematocrit goes up to 90%. The immature blood cells disappear largely, and the animals gain weight and live happily.

So I think this approach is much more realistic for application to humans in bone marrow diseases. One prediction would be, because we know how to do bone marrow transplantation, this might be one of the first diseases being used by this methodology. In Parkinson's, we know from clinical trials, when you introduce into the Parkinson patient neurons, which I derive from aborted fetuses, they can improve substantially the movement disorder.

The problem is, you need 12 fetuses [INAUDIBLE] for one patient, so it's almost impossible to coordinate and quality control is difficult. So a nuclear transfer came online. People thought therapeutic cloning might be a way, but, of course, as I said, it is not really an option. So the question is, can iPS cells do this?

That's what we wanted to test, so we made iPS cells, differentiated those to neural precursors, which can make neurons and glia. And this was tested in the so-called Parkinson rat. This is a model which is produced by injecting a poison into one side of the brain-- 6-OHDA, which kills all dopaminergic neurons on the site. And you then stimulate the animals with amphetamine. They can stimulate on one side but not on the other-- they're not cells. So they begin to rotate.

And so the experiment is, then, to inject your test cells into the lesion and ask the question, do they rotate less? So, clearly, this has nothing to do with Parkinson. Because this only tells you in vivo testing of these cells.

So in the experiment, as you can see here, five rats-- they rotate at 1,200 RPM per 90 minutes for this one here. And four out of five really substantially [? reduced ?] the rotation after transplantation. The blue one actually rotated the other way, then, so there was an imbalance from the injected cells. And four out of four, after eight weeks, did this.

So clearly, again, these cells act as if they act in vivo. So where do we want to end up? I think we want to end up in a clinical setting where you take cells from a patient and derive them under easy conditions for customized therapy.

So let's say, a skin biopsy or a blood sample and in vitro generate ES or iPS cells, and then the cells you want to inject into this [? everything ?] in the Petri dish. So this is actually a very old slide, before iPS were invented. And I thought, always, this was a key issue-- how do we learn how the egg does it and do it without the egg?

I think this is [INAUDIBLE] with the iPS technology. This has changed it. I think this is a problem which needs attention. I believe this scenario, which really looked like a fantasy a few years ago, has become closer to reality, that we, indeed, can now study complex human diseases-- potentially in the Petri dish-- and eventually contribute to therapy. So I think it's fascinating. I feel very privileged to have been able to be in this field and contribute at least something to it.

But I want to finish with another thought. I think getting an award like this is really not half-deserved, or much less than half-deserved. I think getting such an award really is based on all the work of people who worked in my laboratory.

And I have been really, very lucky to have terrific people over the decades in my lab, who did the experiments I talked about. So really they, I think, deserve this much more than I do. So let me introduce some of these people at the end.

These are three individuals who run our mouse colony, which is very complex. Jessie Dausman was the first hire in my career. I hired her in October 1973-- so, what, 37 years ago? She really was, with me, making the first transgenic mice, and she taught how you treat mice and how you manipulate mice to probably everybody in my lab. [INAUDIBLE]

And, actually, as it happens, she's leaving this month. This is her last month. She is retiring. And we were lucky to recruit Kibibi Ganz to take her position. And Ruth Flannery is the other person who is really, absolutely essential for our mouse colony. I hired her, actually, yesterday, 25 years ago, and she is keeping the mouse colony from degenerating into chaos.

It's extremely important to have such dedicated people. But we're not depending only on the mouse colony, but also on our tissue culture, which is very complex. And these are Styliani Markoulaki, Carrie Garrett-Engele and Raaji Alagappan, the other three who really prevent chaos in the lab. Without them, I think, it would be more chaos, and I think it functions because of them.

And they're also helped by two other technicians-- Ping Xu and Dongdong Fu. And nothing of this would help me without Gerry Kemske, my administrative assistant. She really organizes me, organizes the lab, and, without her, I would have been bankrupt a long time ago. So I think Gerry is a key person here.

It was unfortunate that I couldn't really talk about the work of many individuals in the lab. I only mentioned a few. They do terrific work. I didn't have time, so I only want to show a picture of the lab as of last week, when we had our group meeting. This is the group that does all these experiments. I didn't have the chance to mention them.

But I want to mention other people who have left my laboratory, who set up the reprogramming field and went on to their own careers-- Laurie Boyer at MIT, Robert Blelloch at UCSF, Kevin Eggan, Konrad Hochedlinger, Alex Meissner-- they were all students. They all ended up at Harvard. Kathrin Plath at UCLA, Marius Wernig at Stanford, and then, soon, Jacob Hanna will be at the Weizmann Institute in Israel, and Chris Lengner will soon be at the University of Pennsylvania.

But really, this is not enough to do these experiments. I think you need an environment. And the environment, in my opinion, in MIT is unmatched. The Whitehead has collaboration with everybody, but particularly with Rick Young, a friend of long collaboration, with Hidde Ploegh, with Susan Lindquist on the neurodegenerative diseases, with David Page on development, and Bob Weinberg on cancer aspects.

And the biology department [INAUDIBLE] a few-- Tyler Jacks, Phillip Sharp, Alexander van Oudenaarden-- some really great collaboration, Laurie Boyer, and Graham Walker. And in other departments, really, very satisfying collaborations with Bob Langer and Dan Anderson on learning how to grow cells in different conditions by using the engineering approach they have, and even with electric engineering with Joel Voldman. We had a productive collaboration.

And finally, let me just finish with some [? outside ?] [? acknowledgment, ?] which I mentioned. Some outside collaborators-- with Sangamo, which [? are ?] great for the zinc finger approach for us, Tim Townes on the sickle cell anemia model, and Ole Isacson on this Parkinson model. Thank you very much, again, for coming and for this [INAUDIBLE]


MATTHEWS: I want to thank Professor Jaenisch myself for giving an inspirational lecture. And he has said he'd be willing to answer a few questions if there are questions in the audience. Could you move up to the microphone there? The lady in the blue.

AUDIENCE: [INAUDIBLE] I was thinking about your stuff on neurological diseases such as ALS. The interaction between glia and neurons is so tight, especially in myelinization, et cetera, et cetera. How do you, with your stem cells, try to deal with cell-cell interactions, that they give you a disease rather than a single cell?

JAENISCH: I think that's a very important consideration. A disease which is what we call, "cell-autonomous," only the cell itself, is easier to study than when several cell types interact. And ALS is this example.

And, indeed, in the mouse model that Kevin Eggan did, he found that the glia cells are the key part. If they come from the ALS donor, then the neurons will not do that well. So I think you can set up systems in the Petri dish where cells interact. I think that is very important for many diseases, and we'd be doing also some other diseases which have exactly this as a focus.

JAENISCH: Another question over here, please.

AUDIENCE: Thank you for the lecture, Professor Jaenisch. It was interesting to hear some of the more creative claims about stem cells. There was actually a lecture at the nearby Broad Institute on diabetes as part of the Midsummer Nights' Science lecture series. One of the topics was about diabetes and the role of beta cells in that condition.

And the presenter mentioned that there are three major approaches to the disease right now. One of them is, of course, converting a stem cell into these beta cells. Another approach is to convert alpha cells into beta cells, and then, finally, the third approach is to prevent cell apotheosis. Could you comment on the first approach?

JAENISCH: Yes. I think these are all really interesting approaches. Let me come to the last one-- namely, converting exocrine cells to endocrine cells-- so, other cells to beta cells-- which has been achieved in the animal by [INAUDIBLE] laboratory by introducing some key genes. It's a very interesting experiment which says, the epigenetic state-- the iPS approach takes these somatic cells back to the beginning, and now, you have a pluripotent cell that can do everything. This-- it was called transdifferentiation-- can do it horizontally.

Now, it hasn't worked for many cell types. This is one example. Of course, the alpha and the beta cells are very close developmentally, so it was not such a big hump. It was a very interesting experiment. People are doing this now for other cell types. And I think that's a very important approach.

So I agree with you. These are very different approaches. We don't know which, eventually for the clinic, would be the most suitable. I think they have to be pursued-- all of them. And we have to figure out later what's the best for a given situation.

AUDIENCE: And I guess your group is probably hard at work at providing the groundwork for that. So, thank you.

JAENISCH: Sure, I mean, behind these experiments are basic biological questions. Namely, what's the difference, epigenetically, between two cell types, and how do you convert one to the other? And this, of course, is of great interest to us.

MATTHEWS: Another question?

AUDIENCE: Hi. Thank you. My first question is-- initially, you started about epigenetic program's importance in the embryonic development in the initial [? couple ?] [? of ?] cell division. Yamanaka's experiment tells us that just four transcription factors are sufficient to entirely reprogram epigenetic control. How is that possible?

JAENISCH: This is a misconception, so I didn't go into the mechanism. Some people think you add these factors in there like a switch, and your embryonic-- no. That's not how it works. It's much more complex.

What happens is-- how we see it now-- that these four transcription factors initiate a very long process which involves much remodelling of genes, epigenetic reprogramming, and many cell divisions. So DNA has to replicate. The DNA replicates. That's a chance for slowly modifying and reprogramming the key genes.

What you have to do to reprogram a cell-- you have to activate the endogenous pluripotency genes-- the Oct4, Nanog, and Sox2. That's the key. So they have to be really woken up from a dormant state. Right? Demethylate, and so on and so forth.

This doesn't go like a switch. It takes many cell divisions. So this is a bit of a misconception, how that works, but many people have that. It's really that they initiate a long process, which involves many stochastic steps.

AUDIENCE: And my second question is-- embryonic stem cell has the potential to [INAUDIBLE] to all different lineages, right? So, in culture, when we grow them, why do we fail to develop developing embryo?

JAENISCH: Yes. Well, the egg has some unique features. So the egg is a very big cell. It's a thousand-fold bigger than the fibroblasts you start with. And during egg maturation, the mother deposits many, many components which we don't know, really, and which are key for making an embryo.

So you can make mice from embryonic stem cells, but only when you put them in an already-developing embryo, as I showed in the blastocysts. Alone, they can never do this. These cells are small cells. The only thing they can do is divide very rapidly or differentiate. That cannot make a mouse. By themselves, it's impossible. So we need the egg, still, for that.

MATTHEWS: I don't see any further questions. And, before I thank everyone for coming, let me remind you that there is a reception immediately following the lecture, in the Bush Room, which is 10-105. It's almost directly below this room. So I hope to see many of you there. You can chat with Professor Jaenisch, and let's thank him again for a wonderful lecture.