Symposium 3: The Genetic Revolution speakers: Eric Lander, Christopher A. Walsh, Pamela Sklar, Catherine Dulac. Introductions by H. Robert Horvitz, date: May 23, 2011 topics: Epigenetics, Schizophrenia, Cognition, Genetics, Neuroscience

Download 101.16 Kb.
Date conversion19.04.2017
Size101.16 Kb.
  1   2   3

SYMPOSIUM 3: The Genetic Revolution

SPEAKERS: Eric Lander, Christopher A. Walsh, Pamela Sklar, Catherine Dulac. Introductions by H. Robert Horvitz,

DATE: May 23, 2011

TOPICS: Epigenetics, Schizophrenia, Cognition, Genetics, Neuroscience

H. Robert Horvitz:

Well, what a day. And yet another tough act to follow, Francis. I want to start by thanking Patrick Kennedy, Steve Hyman and the other organizers and all who’ve been involved for putting together this spectacular and spectacularly important program. I think all of us here agree that there is an urgent need for advances in the field of neuroscience and I think many of us believe that there is an unprecedented opportunity to make such advances. In this afternoon’s symposium, we’re going to hear about one area with striking opportunities for neuroscience. And that is, genetics. Now Francis has just described some aspects of what this symposium is aptly titled, The Genetic Revolution. And we will expand on Francis’ comments. So, let me back up for a moment. Genetics simply put, is the study of genes. Genes provide the blueprint for how the brain develops and define the components the brain uses to function. Remarkably, genes specify a brain that’s not fixed in its capacity, but rather displays enormous flexibility. Plasticity, as we heard the word earlier today, are the capability of changing structurally and functionally in response to complex and varying environments and experiences. Variations in genes confer differences amongst individuals and brain development and function as well as differences and susceptibilities to brain disorders. So understanding the genetic bases of brain development, brain function and brain dysfunction is crucial for developing treatments for brain disorders. Now, the field of genetics, as I’m sure everybody knows, began really with the discovery of genes by the Austrian monk, Gregor Mendel. Many years later, fast forward, we have DNA as the genetic material, Watson and Crick with the double helix and in the 1960’s the elucidation of the genetic code and the central dogma of molecular genetics. DNA makes RNA makes protein. Now, with this knowledge, genetic studies that were primarily focused on simple laboratory organisms have led us today to an understanding of a vast array of basic biological processes. The organisms studied include bacteria and their viruses, single cells yeasts such as are used for brewing beer or baking bread, microscopic worms, tiny fruit flies and others. And one of the major findings of this research, I think, a conclusion that is striking, and to some, I think, was shocking, is something that I refer to as the principle of the universality of genes. Genes and gene pathways are so similar among superficially distinct organisms that what is learned from studies of yeast, worms, flies, often leads to an understanding of human biology and human disease. Now, a little bit earlier today, first Story Landis and then Ted Kennedy, tried to convince me to tell you a personal story, a story that’s basically my story of worms, genes, apoptosis, a word Francis used, and human disease. I’m not giving a talk. I’m giving an introduction, so I don’t think I have time to tell you that story, but what I will tell you is the conclusion that emerges from that story, my story, and also from the story of many of the scientists in this room and many other scientists around the country and around the world. And that conclusion simply put is the following. Basic research is the driver of biomedical knowledge and of medical progress. Period, full stop.
Now, back to the specifics of genetics. Until recently, genetic studies involved single genes or small sets of interacting genes. Today, however, genetics has, in many cases, evolved into genomics, by which is meant the parallel study of essentially all of the genes and in organisms’ genetic material – its genome. And what’s made this revolution possible is knowledge obtained from single gene studies combined with new technologies. Technologies that, in fact in many cases, had been developed using these very simple organisms that I was alluding to. So today, vast quantities of high resolution genetic data can be generated and, as Francis also alluded to, thanks to advances in the world of computation, analyzed in new ways. And these are the achievements that have opened the door to today’s genetic revolution. New technologies are making possible striking advances in the understanding of how, when and where genes act and in how gene dysfunction can either cause a specific disease or cause a pre-disposition to a specific disease. So in this symposium, we have four speakers who will discuss four aspects of this genetic revolution. We’ll start with Eric Lander who will describe some of the genomic technologies of the genetic revolution, how these technologies have made possible novel analyses of human genes and their effects on human disease and how this knowledge had impacted therapy. Mostly, Eric will be talking about disorders that are prominent primarily non-neuroscience, non-brain in nature. We will then turn to how these technologies are being applied to brain disorders. Christ Walsh will discuss some brain disorders that are associated with single gene defects and show basically how Mendel got it right for such disorders. For example, microcephaly in which the brain is abnormally small and the patient has severe cognitive disabilities. Pamela Sklar will then turn to schizophrenia, a disorder with a major genetic influence, but for which no single gene plays a predominant role and instead, there are complex interactions involving a multiplicity of genes. And then finally, Catherine Dulac will discuss a mechanism of inheritance that does not involve DNA sequence per se. and this so-called epigenetic inheritance, she will explain, is important both in normal learning and in mental retardation. Each speaker will emphasize the opportunities for and an urgency of a major increase in support for neuroscience and each will present, I think, a brief moonshot vision for how investment today in the genetics of the brain might revolutionize our understanding of the brain and our ability to treat brain disorders. I want to emphasize that there is an enormous amount we do not know echoing again what Francis said earlier. And in my view, for us to effectively attack the horrible brain disorders that afflict humanity, we need to support first and foremost that basic and collaborative research that will lead us to the next stages of the genetic revolution in brain science. So with that, let me introduce our first speaker, Dr. Eric Lander. Eric is the founding director of the Broad Institute at MIT and Harvard, co-chair of President Obama’s Council of Advisors in Science and Technology and on a less significant note, was briefly a post doc in my laboratory many years ago. Eric.

[laughter and applause]
Eric Lander:

All right, I’m told we’re supposed to speak from here, but I’ve got to stand up because I’m too excited. As Bob as said, I am the director of the Broad Institute. And the Broad Institute is a very unusual sort of organization. It is a collaborative entity that stretches across all of Harvard, all of MIT and all of the Harvard affiliated hospitals. It is an example of the sorts of collaborations that people are realizing are now necessary to take on the really important problems in biomedicine. And as Bob has also said, I co-chair President Obama’s Council of Advisors on Science and Technology and in working for a man who reminded us that “Yes We Can,” I think that is an extremely important lesson for today. .And so the two themes I want to take are collaboration and Yes We Can when we put our minds to it. So let me start by stating what should be, but is not usually considered obvious, understanding the basis of a disease is a critical foundation for diagnosis and treatment. That was actually not the working assumption through most of the 20th Century. Through most of the 20th Century, we worked by guess work. Guess work occasionally succeeds, but it’s no way to do business. To really systematically attack disease, we need to know what is fundamentally at its root.

Now, we know in, for example, cancer that this is leading to dramatic expansions in therapeutic possibilities. There are more than $500 new chemical entities being tested in cancer and the vast majority of them, certainly those that anybody has any hope in, are based on molecular understanding of specific defects in specific types of cancer. Tests that go on with respect to breast cancer and lung cancer drive patient care and drive therapeutic development. We know outside of cancer the same is true. Genetic discoveries of the genes that lie at the heart of particular diseases are the driving force in therapeutics. We know that this has not yet penetrated psychiatric disease, for example. This is a graph that shows the number of mechanistically distinct drugs available for major depression, for schizophrenia, and for heart disease in 1950 and in the present. The green bar that shoots all the way up there is heart disease. The two little puny bars that stay the same for sixty years are depression and schizophrenia. It shoots up, the green bar because a tremendous amount was learned about mechanism.

In psychiatric disease, there are great examples of failures where people had hope and drugs failed and it should be said, they had no business to hope because they had no mechanistic basis for the hope. How do we do better? How do we understand the basis of disease. Obama’s already said the universality of genetics. This universality of genetics takes us back to 1911 when an undergraduate student working at Columbia worked out a way to map the genes that underlie particular traits. Alfred Sturtevant showed you didn’t have to know what the gene was to find out where it was. He mapped fruit fly genes in 1911 and it became the tool of geneticists all through the 20th Century. But, you couldn’t do it in humans until, until about 1980 or so. The basic notion was kind of straightforward, just like any other Mendelian trait, going back to Gregor Mendel, you could trace the inheritance of a trait in a family, see who was affected or wasn’t affected. But in order to map what that gene was, you need to have a genetic marker that you could correlate to the inheritance of the trait. In 1980, David Botstein proposed a very simple but powerful idea. That, instead of – as in fruit flies – curly wings and funny bristles and things like that that would be used as the markers, simple spelling differences up and down the chromosomes could serve as our genetic markers. And here in this picture, you could trace the inheritance of, let’s say it’s Huntington’s disease, along with the inheritance of a particular letter at one spot in the human genome. The C that’s being inherited along with the disease. And if you saw for enough families that the disease tracked along with the letter C, not just for this family but enough families, you would know that the disease gene and that spelling difference have to be nearby on the

Eric Lander:

chromosome. That idea led to the mapping of Huntington’s disease. That idea led to the mapping of the gene for cystic fibrosis, the chromosome number 7. It led to the mapping of many other diseases. Now, of course, nearby is only so good. In 1985, I remember the day that David Botstein took me into his office, closed the door and whispered, “The gene for cystic fibrosis is on chromosome 7!” It was really exciting and within a couple of months, there was a genetic marker that was 99% correlated. Only 1% of the time did it recombine, but 1% meant a million letters. A million letters took forever in those days. And folks signed up, including Francis, and they signed up to traverse those million letters and it was a long journey. It took four or five years. It took tens of millions of dollars. It took more than a hundred people working on this, but eventually they got there and it looked like this. A lot of letters. But this is not a trivial bit of letters because that little red rectangle there contains a CTT, and those three letters are deleted in the vast majority of patients who have cystic fibrosis. That’s a cause. What does this gene do? Well, you could do diagnostics on those three letters, but you can do even more. You can toss it into the computer and they did toss it into the computer on their very first paper. And the computer came back and said, “Oh, that gene you’ve just found looks a lot like other genes that encode proteins that sit in the surface of the cell and transport things. Congratulations, you’ve probably found the transporter.” Fantastic. This notion that we could map things in a systematic fashion and get to the heart of a particular disease was exciting. And at the same time, it was a little scary because it took five years and tens of millions of dollars and a hundred people working on simply one disease. This was not workable. That was the point of the Human Genome Project. The point of the Human Genome Project was to say if we collaborated together, if we did something bigger than our own individual lab, yes we could make this possible not to need a whole army, but for individual students to be able to carry out projects like this. It required revising the way we think about how to do science together. Setting some goals. That we would build these genetic maps of spelling differences up and down the chromosomes and these pieces of DNA so we could walk along the chromosome in a sequence and have it all available on your iPod, iPad and your gene lists annotated so you know it was there and that all of this would be freely and immediately available to everybody. It was a different way of doing business. And it changed the way we did our labs. We needed to bring in engineers and computer scientists and chemists along with the biologists and the clinical scientists and create things that look like Henry Ford factories because that’s what it took at the time. And we needed to engage in international collaborations.

The Human Genome Project was six countries working together, twenty centers around the world. Large, small, all devoted to one common mission, of getting all those data freely and immediately available. It produced a rough draft sequence in 2001 and a finished sequence in 2003 and now any biologist who wants to can navigate the human genome and know what nucleotides are there and roughly, pretty good right now, what genes are there. That’s an example of what we can do by working together. Now, around the time of the finishing of the human genome, there was a meeting we had out in California. Francis remembers it well. Where we said, yeah, one human genome, that’s great. But that’s never going to be enough.

Eric Lander:

We’ve got to do this for thousands and tens of thousands and hundreds of thousands of human genomes. And we can’t afford to do this. And at that meeting, there was a discussion of a thousand dollar genome. Having just finished the four hundred million dollar genome, there was a modicum of chutzpah involved in the discussion of a thousand dollar genome, but it’s happening. All sorts of new technologies have come along. Technologies, all different kinds of things that use optics and massively parallel sequencing and all that, and Francis has shown you the version he’s cribbed from our slide at the Broad Institute here.


We all crib these from each other. That one I happened to recognize the curb because it exactly is the cost curve for the Broad Institute. And, in fact, it’s now down to a hundred thousandfold cheaper. Nothing has dropped this fast in the course of a decade. What is the impact of a hundred thousandfold cheaper? Let’s talk about where this is going and where it needs to go. For understanding the basis of disease, for those simple one gene Mendelian disorders like cystic fibrosis, we’re doing okay. Back at the beginning of the Human Genome Project, only about 70 had been identified molecularly. By the time of the finishing of the Human Genome Project, about 1,300, by today about 2,800 or so simple one gene Mendelian disorders. In our catalogues, there’s only another 1,800 or so that we know of. So we’re already on the back nine, so to speak. We’re reaching the point where we’ll have most of those known. But of course, that’s not enough. We really need the common disorders. The heart disease, the diabetes, the stroke, the asthma, the autoimmune diseases and major depression and bipolar disease and schizophrenia and other brain disorders. These are not single gene disorders. They’re multi-gene disorders. Well, the time the Human Genome Project started, we kind of only knew one gene that was involved in these multi-gene disorders, the HLA complex. The time the Human Genome Project was nearing its end, we only knew about two dozen or so. It was clear we needed other techniques for his. And so here an idea began floating in the mid 90’s. Hey look, there’s only a certain amount of common genetic variation in the population. Only about ten million or twenty million common genetic variations in the whole human population, depending on exactly how you keep score of what’s common. Why don’t we just find them all, all ten or twenty million of them? And then let’s correlate which ones are at higher frequency in individuals with schizophrenia or with early heart disease. The only problem with this of course is, we only knew a couple thousand. We wanted to get ten or twenty million. We needed to test them in thousands of people. Testing tens of millions of things in thousands of people is tens of billions of genotypes, which at the time were done one at a time by graduate students who, on the whole, objected to doing ten billion things.


And yet, you put one foot in front of the other and, yes we can move ahead in collaboration. Here’s what happened. We went from having only a couple thousand variations to within a few years having a million variations to, on this slide, last year, twenty million variations and currently, this is now 2011, about thirty-eight million genetic variations are in the catalog right now.

Eric Lander:

We went from thinking we had to test them all individually to recognizing that they were locally correlated with each other so you could test the subset and they would proxy for the rest of them. We went from doing them one at a time, to ten at a time, to a hundred at a time, to a thousand at a time, to a million at a time on DNA chips, to soon, five million at a time. And what it meant was as those tools became available around 2006, we went from about one measly discovery of a gene involved in common disease each year to 2005, a few, 2006, 2007, today, more than – it’s actually twelve hundred different genetic variants that have been associated with common genetic diseases. What is it doing for diseases? For diseases like Crohn’s disease and ulcerative colitis, we know a hundred genes involved in these two diseases and have rewritten the textbooks about them. They fall into about ten pathways. Pathways involved in autophagy, how a cell digests itself. Pathways like innate immunity. Pathways involving the signaling of the IL23 receptor, whatever that means for those of you who aren’t scientists. The point is, these things fall into pathways. They make – people have begun to make mice that have these mutations. They show the phenotypes and one can begin to do drug development because one has been able to lift the hood and look what’s wrong under the hood. It isn’t just Crohn’s disease. Age-related macular degeneration we now know is a disease of the alternative compliment pathway. Sickle cell. People knew for decades that with sickle cell anemia, if you could only turn back on the fetal hemoglobin, it would help ameliorate the disease. But no one knew the control of the fetal hemoglobin system. But this is revealed in a particular gene. Cholesterol, early heart attack, new genes have been found that affect that pathway. Cancer. The same sort of things are going on. Not the inherited genes that cause cancer, but the mutations that occur in your own cells during your lifetime that cause cancer. We went from 12 genes, to 80 genes to 240 genes involving all sorts of discoveries that have led to new drug targets. The EGFR in lung cancer, the ALK gene in lung cancer, a new gene will come on the market soon directed against the gene called BRAF in skin cancer. But not only that, we recognized around 2003, 2004 that we were going to need a more systematic attack. So several of us who served on the advisory board for the National Cancer Institute put together a plan to create a systematic attack. A cancer genome atlas program. That was launched about three or four years ago as a pilot project. It’s now successfully launched as a full scale project. The world has chimed in behind it. As an international cancer genome consortium. And we now see a plan ahead of us where for every important type of cancer, thousands and thousands of cases will be sequenced and analyzed. I believe we will have, in a five to seven year time frame, a textbook of what is wrong is most types of cancer. That we’ll provide a road map for how to proceed with therapeutics. It’s still a long path, but imagine doing the path without it.

What about the genetics of psychiatric disease? That’s what we’re here on this panel to discuss. What about it? Well, some people say it’s hard. It might take a lot of patience. It might, might be a lot of work. And it’s easy to get discouraged. If you want to be discouraged, let me show you the early results from the genetics of height. The first ten thousand patients that were looked at, the number of genes convincingly shown associated with it was zero.

Eric Lander:

For Crohn’s disease, two. And yet, as people increase the sample size from 1x to 2x to 3x to 9x to 18x, the number of genes has begun to increase and increase and increase. More than 180 genes related to height. Fifty-one here for Crohn’s. That’s actually out of date. I know it to be about 71 for Crohn’s here. Where are we with schizophrenia? Well, 1, 2, I happen to know that Pam will tell you unpublished results about 6. Do we have the energy? Do we have the commitment? Do we have the funding? Do we have the collaborative will to go up this curve and lay out the road map for schizophrenia and for bipolar disease? I certainly hope so. I certainly hope so. We’ve seen again and again and again, that when the intellectual ideas are there, the technology is there and the commitment is there, it is time to make comprehensive attacks. It worked for the Genome Project. It worked for finding all the common genetic variation. It’s working in cancer. It’s worked through the technologies to sequence. It will work, it can work, it must work for psychiatric disease. I give you to the panel who really works on psychiatric disease. I’m merely the setup man for them. But, it is a remarkable time we are at right now and if we fail, it will not be because the science is not ready. It will be because we will not have been able to marshall the will, the political will, the financial support to do it. And I don’t think any of us in the room are prepared to see that happen. Thanks.


H. Robert Horvitz:

Dr. Christopher Walsh will follow those remarks. Chris is a professor at the Harvard Medical School. And investigator at the Howard Hughes Medical Institute and chief of the division of genetics at Children’s Hospital in Boston and he’ll talk to us about the topic, Learning from Mendelian Disorders, Genes and Cognition. Chris?

Christopher A. Walsh:

Well thanks very much Bob, and I’d like to thank Patrick Kennedy and the organizers for giving me a chance to speak here today. And I’d like to thank them not quite as much for having me speak right after Eric, brilliant and breathtaking tour de force about where we’ve been and where we’re going in human genetics. But I’d like to take the opportunity to extend his comments and tell you a little bit about how these advances in genetics can and have affected families and children with developmental disorders that affect the brain. So, genetics is all about the story of, is all about the study of families. And, I guess, my own interest in genetics started by looking at my own Irish Catholic family and seeing how dyslexia segregates in our family and how my brother has it and how he always grew up reversing his letters and we could see the same trait in my uncle and, then later, we saw it in some of my brother’s kids. And as I went through school, I kept wondering how, how do genes relate then, to these cognitive functions? How can there be a gene that causes you to write your letters backwards? And so, that’s one of the biggest challenges, really, in neuroscience, is how do we map genes onto cognition? We think of our cognitive traits, things like language, things like musical talents, things like scientific ability, how you write your letters and yet, in some sense, as Bob pointed out, the brain is put together as a set of genes that are expressed during embryonic development and those genes then are somehow essential for putting together our brain, the way it is and we know that our brain has tremendous plasticity. But how come my brother still writes his letters backwards? And so why are some things plastic and other things not plastic?

  1   2   3

The database is protected by copyright © 2017
send message

    Main page