Technology Day 1999 - "The Human Body: Emerging Medical Science and Technology"

Search transcript...

HECHT: For those of you who don't know me, I'm Bill Hecht. I know most of you. I'm a member of the class of '61. And I'll explain the red coat in a moment. I would like to say that the job of Executive Vice-President of the Alumni Association has aged me so much that I deserve one of these. But not quite true.

Many years ago, the class of 1924 elected me an honorary member. And unfortunately, none of the 30 survivors of that class are with us today. And therefore it's incumbent upon me to represent that class. So I may be the spryest member of the 75th reunion class here.

Although when they're here, I have to be very careful. They have been known to take me to the Faculty Club and produce the fact that they're much better survivors at some things than I am. We do. Okay. Then I'm in real trouble, because I'm supposed to wear this coat with a freshman tie all the time.

My principal job is to introduce the President of MIT. But I wanted to tell you all that the person who we owe a great debt to is not with us today. That's Phil Sharp. Phil, as you know, is currently and soon to not be chairman of the biology department here at MIT, a Nobel Laureate, and an extraordinary man, who called us several months ago after actually helping us put this program together, and asked permission not to be here because he's receiving an honorary degree in Sweden. And we gave him permission not to be here.

The other thing I want to say is that Bob Weinberg, distinguished member of our faculty, fellow alumnus, has a family issue and will not be at the question and answer period, but will join us and speak as well.

My job is really to call you to order and introduce my favorite honorary alumnus and his favorite bride-- his only bride, actually-- also an honorary alumna. And that's Chuck and Becky Vest. And Chuck will introduce the morning program. Let's welcome Chuck Vest.

[APPLAUSE]

VEST: Well I'm not as fast on my feet as our commencement speakers yesterday, Bill. But I have to tell you a funny story, because at lunch yesterday, with Tom and Ray Magliozzi-- Click and Clack-- our great commencement speakers, were there with all their families toward the back end of the room. And our chairman, Alex d'Arbeloff, was introducing people. And he was of course reading from a long list of their family who were present with them. And he introduced Tom's wife. And apparently, couldn't hear too well in the room, because when he introduced, two women stood up in the back.

And quite appropriate, quite appropriately, Alex looked back and he says, Tom, I don't understand. I announced your wife was here, and two women stood up. What's going on? Without a millisecond delay, Tom said, the other two couldn't come today.

But I've only got one. I guarantee. But it really is a great pleasure to welcome you to what I know is going to be a really extraordinary morning discussion led by four really extraordinary people whom I will introduce in just a few moments. But before doing so, I want to say that this really is a terrifically exciting time at MIT. And I hope you're going to get a little sense of that during your day or two with us here.

It's exciting. It's measured by the achievement of our undergraduates, by the strength of the new class that we have just admitted. I could give you a whole lecture on that. And actually I'm going to. But no this morning.

The significant change in growth of our physical campus. I think you're really going to be quite astounded at what you're going to see as literally a transformation in much of MIT's campus over the coming decade. The continued diversification of our educational and research agendas and their focus on the future, and our interaction with the national and global communities that we serve in so many ways.

The academic year that has just ended has been challenging, as always, and yet very, very fruitful. Early in the spring, for example, we had in a single 24-hour period visits to our campus by US Secretary of Energy Bill Richardson, Bill Gates, and Prime Minister Zhu Rongji of China.

In many ways, representing the breadth of what we do here, starting with fundamental science, because Bill Richardson was here to visit the Bates Laboratory. Bill Gates, of course, representing the future of computing and the role of entrepreneurship in our world today. And Premier Zhu Rongji, I think, very symbolic of the fact that MIT is deeply engaged on a global scale, and indeed, provides an important world platform for leaders.

Granted, of course, that the roster of campus visitors in those two days were a little bit more high profile than usual. But frankly, not by much. I could easily have sighted any other month in the past year to illustrate that 1999, MIT, as it always has been, an exciting and engaged institution.

During that same month, for example, just a few of the research accomplishments that were announced I might point to. Professor Elfatih Eltahir in civil and environmental engineering reported that records of the Nile River's height can be used to put recent occurrences of El Nino into an interesting historical perspective, yet another contribution to our understanding of the changes in global weather patterns.

Professor Maria Zuber, earth atmospheric and planetary science and three of her MIT colleagues were preparing to unveil a precise new topographic map on the surface of Mars that you probably have seen in the newspapers recently, based on data obtained by instruments aboard NASA's Mars Global Surveyor spacecraft.

And at the Harvard MIT division of Health Sciences and Technology-- HST-- graduate student Jennifer Elisseeff reported that her research team had developed an injectable liquid cartilage that may someday be used in a wide variety of orthopedic and plastic surgical procedures and therapies.

Today's program is going to focus on one of the most important areas at MIT-- the interaction of science, engineering, and medicine to produce extraordinary advances in our understanding of the human body. Before we begin, however, I'd like to take just a few moments to remind you that this immensely valuable field of inquiry is just one facet of the larger scientific and engineering revolution being played out here.

Biology and medicine are understandably a focus of a great deal of our attention today. And it will be even more so as we go forward. But no one at MIT who has been at MIT for long will recognize immediately that advances in these fields are closely interrelated with advances across a broad range of scientific and technological disciplines, all of which interweave together.

So before we delve specifically into the worlds of biotechnology, medicine, and health care, I'd like to take just a few moments to give a slightly larger context of MIT's current agenda.

As my colleagues and I look to the world ahead, I think it's clear that the end of science is nowhere in sight, despite catchy book titles of recent vintage. We are under the search for gravity waves. We're looking for anti-matter galaxies. We're moving ahead in the exciting fields of genomics, and neuroscience, and cognition, and learning. Quantum computing is beginning to come onto the horizon here at the Institute. And I suggest you don't really get me started on my excitement about these and so many other things on the horizon or we won't get to the main program.

Innovation. Looking for innovation and technology and business and organization is clearly the key theme of what is going to be driving economies, whether it be of cities, states, regions, nations, the whole globe.

New integrative approaches to large complex systems and problems and structures are going to have to be developed to attack such serious problems as global scale manufacturing, the issues of our global environment, and so forth. Research universities are increasingly going to be just about the only source of fundamental research and scholarship in the United States. That's been our role for 40 years. But folks, we're pretty much the only game in town today when it comes to truly basic science.

A new paradigm of research university is about to emerge. And I really believe that MIT is at the cutting edge of this. One that plays much more importantly together as partners with industry and with government to work on these complex issues in an objective and measured way.

Of course, it's a future in which the ubiquitous, instantaneous flow of information in digital form is just permeating everything-- the way we live, the way we work, the way we learn, the way we entertain ourselves. And that, of course, is going to have profound effect on education. But nonetheless, I deeply believe, as I know most of my colleagues at MIT do, that despite this, the residential campus experience is still going to be the most important way of educating the best and brightest young men and women. There's just nothing that replaces this incredible magic and synergy of bringing these bright young 18, 19, 20-year-old people together to learn in a very intense academic environment.

The security and quality of our lives is going to increasingly depend on building a coherent, productive society, one formed hopefully around common values, yet one that also revels in the rich diversity of its membership. We are going to need to educate new leaders with abilities, values, and desire to apply the knowledge they develop here, to apply it wisely and creatively to the betterment of the human condition.

So as we look forward, we know that MIT must continue to be a place that attracts the best and brightest-- the best and brightest students, the best and brightest faculty, the best and brightest staff to assist them in doing their work and carrying out their mission. We must be committed to excellence, of course, as we always have been.

But MIT also must distinguish itself by thriving on change. This is not an age in which staying steady, staying still, is going to be the most important thing. We must lead, as I've already said, in developing these new, more integrative modes of scholarship and learning and action. We must be even more adept than we already are at bringing together industry, government, and academia to solve major problems facing the world.

And in order to do all these things, we're absolutely convinced that we must have a strong presence of engagement with the arts, the humanities, and the social sciences. We must expand our technological and organizational capabilities, but also be very much attuned and concerned about the moral and ethical issues that the rapid progress of technology and its effect on our lives is going to have.

And finally, we continue to follow a very important philosophy. It says that MIT is first and foremost an American institution, born and bred and developed here in the United States and by the United States, but in order in the 21st century to be the very best possible university, focused largely on science and engineering and management. We must be engaged globally. We must have a cosmopolitan student body and faculty. And we must be engaged with our colleagues around the world.

As always, MIT's pursuit of this vision will be ably shaped and guided by a brilliant and talented faculty. And the best way to demonstrate the continuing excitement and rich promise of MIT's future is just that, to let the faculty speak for themselves.

Our four colleagues who will lead this morning's discussion are truly extraordinary in every dimension-- in brilliance and accomplishment, and contribution to human welfare, as good and engaged people, and as teachers and mentors for the next generation. They really make us all proud to be part of MIT.

Now in order to move forward efficiently, I'm simply going to introduce all of our speakers now, and then turn the platform over to them to guide the discussion for the rest of the morning.

Let me begin with Bob Langer. Robert S. Langer Jr, class of 1974 as a Kenneth J. Germeshausen, Professor of Chemical and Biomedical Engineering. The scope of Bob's interests and achievement ranges so widely that he is the only active member of the US National Academies of Science and Engineering, and the Institute of Medicine.

A pioneering researcher in bioengineering techniques, his work has led directly to numerous improvements in medical care, including the development of implantable polymer wafers that release time-controlled doses of medication to treat tumors and other localized illnesses.

He has also successfully developed artificial skin therapies to help burn victims. More recently, he and MIT colleagues, Professor Michael Cima and John Santini, have announced the development of the new microchip that can be programmed to release specific doses of multiple medications. This so-called pharmacy on a chip as a wide variety of potential applications, and has already garnered considerable national and international interest.

Bob's work has already yielded some 350 patents. And his extraordinary productivity has earned him a Lemelson MIT Prize for Excellence in Invention and Innovation.

Bob Weinberg. Robert A. Weinberg, class of 1964. The Daniel K. Ludwig professor of cancer research at MIT. He's a founding member of the Whitehead Institute for Biomedical Research.

A leading specialist in the study of the genetic basis of cancer, professor Weinberg has made a number of major advances in understanding the genetic mechanisms to control or inhibit cell growth. Bob's internationally acclaimed work is designed to lay the foundation for a new generation of cancer therapies. In the evolving effort to understand and control human cancer, his discoveries have marked a crucial watershed, and earned him an enduring place of honor among cancer researchers.

He also has just received another enduring place of honor here at MIT. Just three weeks ago, he received the Killian Faculty Achievement Award-- the highest award bestowed by our faculty on one of its colleagues. And also, if I might editorialize a little bit, something I value very deeply about Bob is that he does something that is incredibly important. He is a remarkable communicator of the science of cancer to the lay public, something that I just cannot overstate how important the kind of writing he does for a more lay-oriented audience is.

David Page. Dr. David C. Page is professor of biology, Associate Investigator at the Howard Hughes Medical Institute, and an adjunct faculty member of the Harvard MIT Division of Health Science and Technology. Like Bob Weinberg, David is also a member of the Whitehead Institute where he chairs a task force on genetic testing, privacy, and public policy.

As an adviser to numerous governmental bodies, he is one of those rare and invaluable scientists who combine first-rate research with genuine public leadership on science-related policy issues. His research is focused on mapping the Y chromosome as a promising avenue toward understanding and treating a number of chromosome-based disorders, including infertility. In recognition of his achievement in biomedical science, David received the 1997 Amory prize from the American Academy of Arts and Sciences.

Martha Constantine-Patton. All of us at MIT are absolutely delighted that the next speaker in this sequence-- Martha Constantine-Patton-- will be joining MIT's biology faculty as of July 1. She is one of the nation's leading researchers and teachers in neuroscience, and has received among many other awards and honors, the Matilda [INAUDIBLE] Award in Neuroscience, The Society of Neuroscience Young Investigator Award, and received a John Simon Guggenheim fellowship.

Currently a professor of molecular, cellular, and developmental biology at Yale, doctor Constantine-Patton has served as director of Yale's interdepartmental neuroscience training program. Under the auspices of the National Institute of Mental Health, she has also served as director of two different programs in neuroscience training and behavioral science and studies-- one at Yale, and the other at the Woods Hole Marine Biological Laboratory.

We are honored that she has been able to come up to Cambridge to join us for Tech Day in advance of her formal arrival on campus. So it's my pleasure now to simply turn the stage over to my colleagues. I'm going to turn the stage over to Bob Weinberg. I just made a decision. Bob, come on up.

[APPLAUSE]

WEINBERG: Thank you, Chuck, for that kind introduction. I'm most pleased and flattered to be here on this occasion of my own 35th reunion. I've been here now almost 40 years. And I must say when I first came, biology was only a minor variation on the MIT theme. It was a relatively minor part of MIT's research and teaching agenda. And it's grown enormously since then, to the extent that the MIT Department of Biology is now amongst the top five in this country, and likely in the world, which is an extraordinary testimonial to how things change here in no small part due to the farsightedness of some of our leaders.

I think, in fact, that this enormous expansion is due to two separate factors. First came the discovery of the DNA double helix in 1953 by Watson and Crick. Watson and Crick are still around. They're still alive and kicking and active in the field of biology. And I imagine 200 or 300 years from now, one will look back on this era in which we live as having been as important in terms of its watershed effects on science as was, for example, the discovery of the laws of physics by Isaac Newton 300 years ago.

So we live in an era which really is epoch-making in terms of the opening up the explosion of a new field of science. I'm very impressed by the fact that we live in this time of Watson and Crick. Although I must say most people are much more impressed by the fact of all the work here at MIT goes on the same city as does the garage work of Tom and Ray Magliozzi-- Click and Clack, the Tappet brothers. That's for them, ultimately, a more impressing fact.

In the early 1960s, there was the recruitment of a large number of faculty to the biology department and related areas. And this has resulted in this rich harvest of research, some of which I wish to discuss now, and by implication, work of other laboratories at MIT.

My focus this morning is on how we have come to understand the origin of human cancer. That is, what provokes this disease, and what prospects does this give us in the long run for understanding how to control it.

And in order to elaborate on that theme, I want to begin by discussing one important lesson we've learned. And that lesson, which has been long in coming, is that cancer is not an inevitable consequence of the human condition, but rather that the incidence of cancer occurs at vastly different rates in different sub-populations of humanity. And in saying that, I refer to begin to the discovery in the end of the 18th century by a London physician that chimney sweeps who had worked in their youth--

AUDIENCE: [INAUDIBLE] can't hear.

WEINBERG: Oh. They promised me-- are we okay now?

AUDIENCE: No.

WEINBERG: Is that right?

[LAUGHTER]

I was about to swallow it. Okay. That men who had worked as chimney sweeps in their youths in the soot of the London chimneys came down with cancer of the scrotum at very high rates-- an otherwise very rare kind of malignancy. The first insight-- the first connection-- into the fact that a specific exposure was connected with the onset of a disease, rather than this disease was simply a spontaneous accident.

And over the ensuing 200 years, the evidence of connections between the onsets of certain kinds of cancer and specific exposures has accumulated largely through the work of epidemiologists. Here, for example, is the incidence of cancer of the mouth and throat among women, which we see is concentrated in the Southeastern part of the United States, not because of heredity, but because as you might speculate, of the habit of chewing tobacco in that sub-population.

Or for example, cancer of the liver, which occurs at extraordinarily high rates in certain parts of Africa and Southeast Asia, but occurs relatively rarely, thankfully, in this country, a consequence one now realizes years later, of eating tainted grains and perhaps also of hepatitis B virus infection.

Other evidence from epidemiology comes from studies of lifestyle. For example, the description already in the early 19th century that breast cancer was a plague among the nunneries of Northern Italy, this being due, we now realize in retrospect, to the increased risk of the disease due to the fact that the nuns in the nunneries obviously bore no children, and as a consequence, had a two to threefold increased risk of breast cancer during their lifetimes.

And then there are the discoveries of epidemiology. Here, for example, is the risk of breast cancer in different countries across the world. And what we see is an extraordinary variability from one geographical location to another in the risk of breast cancer-- as much as a 20-fold difference. Not a twofold difference or 20% difference, but a 2,000% difference across the globe in the incidence of breast cancer.

On this graph, it's correlated on the abscissa-- you remember what that is-- with dietary fat intake. Some might conclude that eating more fat gives you more breast cancer. But we will not leap to that conclusion today.

I just want to impress on you the fact that different kinds of human cancer occur with enormously different frequencies for reasons that are not clear. But one thing is clear. These differences are not due to differences in inborn predisposition-- that is to say, in genetics-- because when individuals move from countries of low cancer incidence to countries of high cancer incidence, within a generation or two, their risk of cancer begins to approach to that of the people who've always lived in that country. So it's not as if these people are genetically less susceptible to this disease or, for example, to prostate cancer, which has a remarkably similar kind of spectrum.

Prostate cancer, until recently, has been as much as seven-fold more frequent in this country then it has in Japan, for example. And that's changing now due to changes in Japanese diet. Again, I'm wishing to leave you with the impression that the incidence of most kinds of disease cancer is highly variable, and that to the extent that we are susceptible to these cancers in this country, it's not simply that we're preordained to come down with cancer. There are aspects of our diet and our lifestyle and our environment, very largely diet and lifestyle in terms of smoking, which influence enormously disease incidence.

Here you can see yet another representation of that. Here's the incidence, for example, among males of stomach cancer over the last 60 years in this country. And you see it's gone down dramatically by six-fold. A testimonial to the powers of food preservation, the powers of refrigerating food and the deleterious effects-- at least as argued by some of food preservatives-- which keeps food from going moldy, and therefore protects our stomachs.

So next time you hear food preservatives maligned, think of this decrease in stomach cancer rates over the last 60 years. Here's another change in cancer incidence. This is the change in incidence, of course, of lung cancer, which is by now the greatest killer in this country.

At the beginning of this century, a case of lung cancer was so rare and unique that it would be paraded in front of a medical school class, the professor saying to the students that this was such an interesting disease that they likely would never again see another in their lifetimes.

Here is female cancer incidence over the last 60 years. And these are all age-adjusted incidents to take into account the aging of the population. We see here that, for example, stomach cancer has gone down dramatically. Cervical cancer has gone down due to the pap test. Lung cancer has gone up, and by now exceeds the rate of death in women from breast cancer. And here's an amusing kind of curve. The rate of breast cancer mortality over the last 60 years among women, age adjusted, has changed absolutely not at all.

So the next time you hear about a cancer epidemic in this country, remember that if you subtract away the effects of smoking and tobacco, that the rates of most kinds of cancer in this country have either remained static or gone down. We're not being overwhelmed by an increased cancer risk. And nor is there any evidence that environmental pollution causes any more than a minute fraction of tumors in this population.

Here's another way of depicting the connection between lifestyle and cancer. Here the number of deaths that are caused every year in this country-- this was several years ago-- from smoking. The effects probably of secondhand smoke. And this is a little uncertain.

Here's the number of American deaths in all of World War II or in Vietnam. So the annual death rate for smoking in this country-- much of it due to cancer, not all of it, of course-- is absolutely staggering. Once again, persuading us of something that no longer requires us to be persuaded of-- that is that these various kinds of cancers are not simply a random breakdown of the human body, that their Incidence is encouraged by different ways we treat our bodies.

And that causes us now to focus on the disease of cancer to try to understand what it is biologically. And so doing, we go to a little bit of 701. A little bit of biology. I don't mind doing that since that's the way I earn my living around here.

Here's what a normal tissue looks like. We have something like 30 trillion, 3,000 billion cells in our body. Here's a small slice of that.

And what we see if we look in a normal tissue is that the cells in a normal tissue are well-ordered, that they form a very exquisitely organized architecture. And that this architecture is disrupted when a tumor arises in the body. A tumor, when viewed microscopically, is just seen as a jumble of cells, large numbers of cells whose only objective in life is to make more copies of themselves. These cells in the cancer are not interested in creating finely tuned architecture that functions in one or another tissue.

The cells in this tissue-- in this cancerous tissue-- are only interested in multiplying. That is the only item on their agenda. And if we want to understand how that happens, we have to understand that normal cells grow for a certain finite number of doublings in a normal tissue. And then they stop growing.

And cancer cells, in stark contrast, they continue to multiply without limits, and ultimately yield so many descendants that the mass of descendant cells-- billions and hundreds of billions of cells-- ultimately compromises one or another vital organ in the human body.

Having said that, I want to make one other point about cancer. We might imagine there are normal cells and there are cancer cells in the body. We might imagine that correctly.

And having said that, and looking at the large mass of cells in a human tumor, we might ask how they arise. And there in fact are two countervailing theories as to how they arise.

One theory says that all the cells in the tumor are the descendants of a number or founder cells, each of which cross the boundary from normalcy to malignancy, each of which became independently transformed from a normal cell into a cancer cell. And once transformed, that cell began to multiply exponentially, yielding the large flock of descendant cells.

Mind you, if you see a small human tumor the size of only half an inch by half an inch, that already has a billion cells in it.

The other theory-- the other model-- is that all the cells in the tumor are the descendants of a single founding cell, which began to multiply exponentially. And it's clear now from the detailed studies of all the cells in the human tumor, we now realize that all the cells in the human tumor descend from a single common founding ancestor, that it is a so-called monoclonal population of cells. Monoclonal referring to the fact that all of the cells are in a single flock, a single genetically descended group of cells, all deriving from a common ancestor that crossed the boundary from normalcy into malignancy.

Having said that, one's attention is focused on the behavior of that founding cell, trying to understand what causes it now to behave abnormally, and trying to understand ultimately the environmental factors that have driven or provoked that conversion, that transformation. And in so doing, we focus on a single cell. This is what a normal cell looks like under the microscope. And what we want to do is to focus ultimately on the nucleus of the cell.

This is what an artist's depiction of a cell is. And like artist's depiction of almost everything biological, has not even a vague connection with what a real cell looks like. But here's the nucleus of the cell. And we want to explore the nucleus of the cell, because it's in the nucleus of the cell that the heart and the mind of the cell lie. That's where the brains of the cell are. That's where the control mechanisms are that tell the cell when to grow or not to grow.

And thus we look at the chromosomes within the human cell. In the nucleus of the human cell there are 46 of them. And they contain the genetic information in the form of the double helices of DNA discovered by Watson and Crick, that determine not only how tall we are and what color our eyes are, but also instruct the individual cell as to whether it should grow or not grow. In other words, the individual cell take takes its instruction from the genes that are carried in the chromosomal DNA in the nucleus of the cell.

We also want to focus on the fact that the DNA encodes information in the form of sequences of bases. Here we see them. G-A-C-T-A-G-T. And these sequences of bases, these DNA molecules, like all physical entities in the world, are subject to damage.

And thus when they are damaged-- when a segment of DNA is cut out-- or when the identity of one or another base is changed, for example, going from a c to a t, or from going from a c to a g-- then the information content of the DNA sentence is changed in the same way that dog and dig represent changes in information content. Obviously, in the context of DNA, we call these changes "mutations."

Information has accumulated over the last two decades that within the DNA of cancer cells there lie mutant genes. And these mutant genes are responsible for redirecting, reorchestrating the behavior of the cell, forcing it to behave no longer like a normal cell, but rather like a cancer cell.

And some of that information is inspired by the work of a professor at Berkeley, Bruce Ames, who began to investigate the relationship in a number of chemicals between their ability to cause cancer and independent of that, their ability to damage the genes inside cells. Ames was a proponent of the theory that perhaps the way by which carcinogens-- cancer-causing chemicals-- act is to enter into the cells in the body, damage genes, and thereby trigger the outgrowth of tumors.

And his research revealed the following very provocative outcomes. If he studied the ability of various chemicals to induce mutations in his test, these mutations being represented here on the abscissa, he could find some chemicals which in this plotting were very potent in their ability to cause mutations, to damage the DNA-- the double helices of the DNA. And conversely he found other chemicals which were very weak in their ability to damage DNA.

And independent of these kinds of data, these results, he plotted on the ordinate, on the vertical axis-- I know I don't have to tell you that-- that the ability of these various chemicals to induce tumors in mice and rats. So he compared the mutagenicity-- the ability to damage DNA-- with the tumorigenicity-- their ability to induce cancers. And what he discovered was that chemicals that were extraordinarily potent in their ability to induce cancer and lay on this end of the ordinate were also extraordinarily potent in their ability to induce mutations. And conversely, chemicals-- oh, sorry. Wrong button. After 40 years, still haven't learned.

And chemicals which are very weak in their ability to induce cancer are weak in their ability to induce mutations. And that had led to the thinking that the way by which many carcinogens act is that they enter into cells in the body, attack the chromosomal DNA, damage the sequence of bases in the DNA, thereby generating mutant genes which, once mutated, then begin to release instructions to the cell, which drives the cell into unrelenting cycles of growth and division and replication.

And one could test this kind of model directly and experimentally in the following way. One can take DNA from cancer cells. So here we have cancer cells growing in the Petri dish. The model, the theory, we're trying to test here is the following-- that within the DNA of cancer cells there lie mutant genes which are responsible for the aberrant behavior of these cells. Until now this has been a speculation.

But here I'll show you a real experiment to prove it's true. So one takes-- oh. Wrong button again. Could we go-- there we go. Okay.

So one takes cancer cells growing in the Petri dish and prepares their DNA. And these DNA molecules clearly contain the genes of the cancer cells. And then one can take these DNA molecules from the cancer cell and introduce them into normal cells that are growing in the Petri dish here, and ask the following question. If we put the genes of the cancer cell into these normal cells, will these previously normal cells begin to themselves behave like cancer cells? All that they've received are the genes from the cancer cell.

And in fact, such an experiment succeeds, because if one takes genes and DNA from cancer cells, puts it into normal cells here, some of these normal cells-- previously normal cells-- undergo a conversion, a transformation, that enables them to grow now like cancer cells. Here's what such an experiment looks like under the microscope.

Oh. This is not good. There we go. Okay.

Here are some normal cells growing under the Petri dish. And here are the descendants of a normal cell, which acquired a gene from a cancer cell called an oncogene-- a cancer-causing gene. And now these cells grow abnormally on the Petri dish. And if one plucks these cells out of the Petri dish and injects them into a young mouse, they'll start growing into a tumor.

And such a simple experiment proves unequivocally the point that within the DNA, within the genes of a cancer cell, there lie cancer-causing genes, oncogenes, that can drive normal cells into cycles of malignant growth. And thus the very existence of these genes is no longer a matter of speculation.

One can isolate such genes using modern recombinant DNA technology. Here's the way one would depict such a gene as isolated by DNA cloning. It's a single segment of DNA. It's 6,000 bases long. This is only a minute fraction of the total genetic complexity of the human cell, where there are 3 billion bases of DNA.

And this gene is present both in normal cells and in cancer cells. But the difference between this gene in normal cells and cancer cells is that in cancer cells this oncogene-- in this case, retrieved from the human bladder carcinoma-- has suffered damage. The gene is 6,000 bases long. And the essential difference between the two genes is a single one of these 6,000 bases, as is indicated here.

Here's the sequence of the basis of the normal gene. It goes G-G-C-G-C-C-G-G-C-G-G-T and so forth. And then the cancer-causing gene, retrieved by gene cloning from the cancer cell and sequence-- the sequence has been determined, the sole and exclusive difference that converts the previously normal growth-regulating gene into a cancer-causing gene is a single change of the DNA base from a G to a T.

And I mention this to illustrate the fact that the powers of modern molecular biology can now dissect with extraordinary precision and specificity the molecular changes inside cancer cells that are responsible for their aberrant growth. It's no longer a matter of speculation that mutant genes drive cancerous growth. The mutant form with a T here causes normal cells to undergo malignant transformation.

The normal gene, which resides in all our normal cells and has not suffered mutation, programs normal cell proliferation. And we know how such genes work. For example, the normal gene and the cancer-causing gene both encode the structure of a protein. As is the case with all genes, it's the protein that does the dirty work of the gene.

And this protein works as a molecular switch-- an on/off switch-- which is continually flipping back and forth. When the cell receives a growth stimulatory signal, the switch is turned on. And when the cell no longer receives a signal, the switch is turned off. And that's the way the normal molecular switch works.

But in the molecular switch that derives and is found in the cancer cell, instead of going from an on state to an off state, the molecular switch is left to run in the on state for indefinite periods of time, being unable to shut itself off. And therefore, a vital component of the signaling circuitry of the cell, which normally should be on for a brief moment in time, releasing a pulse of growth stimulatory signals into the cell, is left on at full blast for an extended period of time, driving incessant, unrelenting growth on the part of the cancer cell.

We even know now by looking into the innards of the cell, on the basis of work from hundreds of laboratories, about a complex circuitry, an interrelating cohort of proteins within the cell. Here's the limiting membrane of the cell. And here are proteins which talk to one another, which communicate with one another, and process growth stimulatory signals in the cell, helping the cell to decide whether or not it should grow.

When one of these proteins-- the one I showed you just before, which happens to be located right here-- is disrupted through a mutation in the gene that determines its structure, this circuitry malfunctions. So we now not only understand the identities of many of the genes, but we begin to understand the intercommunicating proteins inside the cell, which operate much like components in a circuit board-- mini computer inside the cell-- which helps to determine whether or not a cell grows.

And when the components of the circuit board-- of the mini computer-- are disrupted through mutations of the gene that determine their structure, then the circuit as a whole is disrupted, and decision-making goes awry.

Here's yet another way of depicting that. And I'm not going to quiz you on the details of this graph. I only mean to impress on you the notion that we now begin to understand the innards of the cell with great precision, the intercommunicating components which enable cells to receive signals, to process these signals, and to converge in a normal cell on the decision as to whether it should grow or not grow.

Knowing that, one can begin to imagine how disruption of one or another critical component of the circuitry can then lead to unrelenting growth. We know as well about yet other kinds of genes. The genes I talked about before are called oncogenes.

And when they are present in mutant form in the cell, they drive unrelenting growth. They are, in effect, operating like accelerator pedals in cars. In the normal cell, the accelerator pedal-- the normal proto-oncogene, the normal version of this gene-- stimulates growth only briefly by being pushed down and then flipping back up.

In the cancer cell, the oncogene, the malfunctioning oncogene, operates like a stuck accelerator stuck to the floor, driving unrelenting cell proliferation, driving the car to to move forward without any limits. There are yet other genes which operate in an exactly opposite, or countervailing fashion, in the cell. These other genes operate like breaking mechanisms, like brake linings, by stopping cell proliferation. And often they're called growth suppressor genes, or tumor suppressor genes.

And we've learned about these growth suppressor genes by studying, among other things, a rare eye tumor called retinoblastoma. One of the interesting things about cancer research is that by studying tumors in one organ, one very often learns lessons about cancers occurring in many other sites throughout the body. So the lessons about studying the rare eye tumor retinoblastoma, a tumor of young kids up to the age of maybe 5 or 6, seen in only one 1 of 20,000 children, is illustrative of the principles that govern the growth of cells throughout the body.

In the case of the retinoblastoma tumor, we know that its origin is do not to be hyperactivation of a growth-promoting gene, but on the contrary, to the inactivation of a growth-suppressing gene. And the consequence is that a gene which is normally 190,000 bases long. Look at the line that's labeled N here and goes from here all the way over to here, in many cancer cells, major parts of the gene are missing. They're eviscerated, they're cut out through mutations.

And as a consequence, this braking mechanism that operates to slow down cell proliferation, is disabled.

I mention that to indicate that we now realize that in most cancer cells, there are at least two kinds of genetic changes of mutant genes in the cell, which result in two distinct changes in control. One are the mutant oncogenes which drive unrelenting proliferation, like the stuck accelerator.

And the other are of the defective tumor suppressor genes, which operates like defective brake linings in the cancer cells. And therefore the forward motion of the cancer cell in terms of its proliferative power, derives from both of these things converging and conspiring. A stuck accelerator pedal and a defective brake lining, in this case being disabled by mutations that destroy the integrity of the gene, that templates the functioning of the braking mechanism

We also learn about these tumor suppressor genes because when they are inherited in mutant form-- in defective form-- from one generation to the next, they are often responsible for heritable kinds of breast cancer-- certain kinds of cancer-- I'm sorry-- in this case. And in the case of other tumor suppressor genes, yet other kinds of human malignancies.

We now believe that between 5% and 10% of human cancer is due to inherited tendencies to one or another kind of disease. Rather than accidents occurring to one's cells in one's genes in one's lifetime, some individuals between, 5% and 105 of cancer patients come down with cancer, not because of the way that they abuse the cells in their body or because of bad luck occurring to their cells and genes during their lifetime, but because they had the bad luck of inheriting a defective gene from one or another parent.

And in such families, one now sees the appearance of cancers at high frequency. Here is a retinoblastoma cancer family. I mentioned before that this cancer otherwise only occurs in one out of 20,000 in the population.

And here you see that in a kindred in which the tumor suppressor gene is inherited and passed on from generation to generation in defective form, the cancer is seen in very high rates. And we know that similar situations obtained in the case of heritable colon cancer, and heritable breast cancer, and a number of other heritable tumors where a portion of the tumors seen in the population are due to an inborn susceptibility.

And finally, I mention one other recent discovery in cancer research that has been most exciting and provocative. And that deals with the ability of cancer cells to proliferate without limits.

What do I mean by that? If one takes a normal cell and puts that cell into a Petri dish and allows it to grow, then that normal cell will proliferate only for a limited number of times in the Petri dish before stopping to grow. If one takes a cancer cell and puts it in a Petri dish, the cancer cell will double forever. And that's really a provocative finding, because it suggests that the normal cell has a built in intrinsic alarm clock that tells the cell it can double for a certain number of times before it must stop growing.

And the cancer cell, on the contrary, has acquired the ability to double without limits. The cancer cell is said to become immortalized. Not the human being, but the endometrial cancer cell, because it can continue to double forever, which apparently is an essential part of its ability to form tumors.

And we now realize the molecular nature of the clocking mechanism that operates in normal cells and inside cancer cells to determine the replicated potential. It turns out that at the ends of chromosomes, there are DNA sequences that are known as telomeres. And we realize that within normal cells, these telomeres grow increasingly short each time a cell goes through a cycle of growth and division.

And when they become critically shortened, these telomeres are no longer able to protect the ends of their chromosomes, which is the normal role of the telomeres. They function in effect like the little shield at the end of our shoelaces, to prevent the chromosomal DNA from frame.

And when normal cells go through increasing cycles of generations of doubling, their telomeres get shorter and shorter, until they finally have shortened so much that they are no longer able to protect the chromosomal DNA. The telomeres and the chromosomes collapse, and the cells die.

Cancer cells, we've learned over the last two or three years, are able to forestall and prevent this telomeric collapse by developing an enzyme called telomerase, that is able to repair and elongates these telomeres, thereby disabling and disrupting this generational clock that prevents normal cells from growing forever.

So even here we've developed an increasing molecular insight into the origins of aberrant cancer cell growth. Here we see the telomeres at the ends of chromosomes. Here they're highlighted in yellow. And here we see a chromosome which has lost its telomeres. It has no yellow on the end of it. And the cell carrying this chromosome is soon going to die by virtue of the fact that the telomere is no longer here at the end of the chromosome to protect the physical integrity of the chromosome, enabling the cell to grow forever.

So having said all that, and wishing, of course, to talk much more. But not doing so, because my time is up. I mean to indicate to us that over the last two decades, from work in a large number of laboratories, we have now begun to understand with increasing precision the molecular abnormalities inside the cancer cells. They are no longer matters of speculation.

20 years ago, the mid 1970s, if you asked a biologist, what's the matter with a cancer cell? Why does it grow so abnormally?

He or she would have shrugged the shoulders and just said, we really don't know. But we now know. It is no longer a mystery. It's no longer a matter of speculation. The average growth properties of cancer cells have now been traced with great precision to malfunctioning molecules inside the cells, to malfunctioning proteins. And these proteins have been traced in turn to mutant genes, which are damaged in the chromosomes of the cancer cell during the development of the tumor. Or mutant genes that may be inherited in mutant form, in defective form, from one or another parent.

And this is most exciting for those of us in cancer research, because it may lead to the following. In fact, I believe it will lead. We have always done this work with the notion, with the faith, that if we understood the origin of the disease, if we understood the motive force behind the aberrant growth of cancer cells, we would be placed in a very advantageous position with respect to our ability to develop in the future new kinds of cures against the disease.

Until now, until very recently, to the extent that we had therapies against tumor cells, they were developed in the face of total ignorance of the molecular aberrations inside the cancer cell. But all that has changed. And not changed minimally, it's changed dramatically.

And so now I feel quite secure and confident predicting that in the next decade of the coming Millennium, there are going to be a whole series of dramatically new types of anticancer therapies that are based on this rich load of this treasure trove of molecular information on the origins of cancer that has been gathered in so many different laboratories over the last 2 and 2 and 1/2 decades.

And on that note, I say thank you. And happy reunion as well to my classmates-- class of '64.

[APPLAUSE]

Thank you, Dr. Weinberg. I have one quick note. If William Stuart is in the audience, could you meet me out in the foyer in just a second? We have an issue with a kid. Not serious.

And now I am pleased to introduce Martha Constantine-Paton.

[APPLAUSE]

PATON: Well, thank you. I should thank Phil for inviting me to participate here today in a little advance of my joining this faculty. It's actually fortuitous that I follow Bob in this presentation, because he had a lot to do with my coming here. When I came up for a job interview, he took me to lunch. And we sat down. And here is a world famous scientist in cancer. And he says to me, in a decade, or maybe two decades, we're going to solve cancer.

This is a person in perhaps the premier biology department in the country. And he said, but we're not close to solving the problems of the brain. And we want to get there. So we would like you to come and join us and help us. That kind of an invitation is very, very hard for a neurobiologist to refuse.

I'm going to talk to you today about developmental plasticity. I'm a biologist who has followed neuroscience in biology departments for over 23 years now. And biologists know about development. Development is change. Plasticity is change.

So what do we mean by developmental plasticity? Well it's simply the potential to change the process of change. And nowhere is this potential more developed in organisms than it is when we talk about the development of this organ that makes us us-- the development of the human brain. Now brain development involves genes, involves more genes than any other organ of the body. But it involves the genes to set the pattern, to set the molecules that interact, and then to open up that pattern to interaction with the environment so that the brains that we use as adults today are the products of our genes and the products of our experience.

And when I talk about developmental plasticity, I talk about how that final product comes about. Now we know a lot about it empirically. The processes that change the brain are iterative processes. The initial experiences change the structure of the brain that then changes the function of the brain that then changes the way the brain responds to different stimuli.

These processes are high, occurring at phenomenal rates in young children. And they are turned down slowly as we mature. Actually, there's a pretty big drop at puberty.

Young children are phenomenal at the process of brain change. As adults, these processes of change are sequestered to regions of our brain that are involved in functions like learning and memory. It would not be good to have a brain that can constantly change all its circuitry and all its wiring throughout our life. But we have sequestered it into select regions to allow us to be learners and adapt throughout our long lives.

Now sometimes this process goes awry. Children are born with genetic effects, defects that don't allow their brains to interact normally with the environment. And sometimes perfectly normal children wind up in what we like to think of in broad terms as toxic environments-- environments that are narrow or restrictive for some reason.

And as this brain plasticity is turned down, some of these effects get locked in and make for maladaptive, abnormal adult behavior. And so it's important for us to try to understand these processes, not only so that we can change the environments in which we raise our children, but also so that we can go in, and as biologists and technologists, tinker with this mechanism, and perhaps reactivate it in regions of the brain, of the mature brain, and thus open up those regions to adapt the plasticity later in life.

So I'm going to paint a picture with a very broad brush today, and give you a sense of what this process of plasticity involves. And I'm going to talk a little bit about mechanism. Quite a focused approach to mechanism. But nevertheless, one that will give you a flavor of what drives those of us who want to understand the biology of the developing brain.

So here we go now. Well, developmental plasticity actually operates at a number of different levels. There are instances-- if I've got this loud enough-- in which deprived early environment leads to permanent functional impairment. Some of the easiest instances of these for us to understand and talk about are ones where a sensory modality has been perturbed early in development.

For example, children who are born with squint-- that is in which one eye muscle is a little loose so that the two eyes can't focus at the same point in visual space-- these children can never use their eyes simultaneously. And so the patterns of activities that the eyes send into their brain are never patterned in time in an appropriate way. And their cerebral cortex responds to that by sequestering the input from the two separate eyes, two completely separate populations of cells, throughout the circuits in the cortex. And that individual grows up never ever able to have true binocular vision, no depth perception, very severe problems in things that we would consider completely normal vision.

But there are much more behaviorally severe consequences of abnormal environments. Children in orphanages that actually don't get the normal stimulation during the first two years of their lives. These children can be adopted, brought into wonderful, wonderful families, and yet they still have throughout their lives some behavioral impairment because of these narrow, deprived early environments.

And in those instances, we are far from understanding the loci of those changes. We can understand the process. And I'll tell you some of the mechanisms. But we don't understand where in the brain those disruptions have occurred.

Over the course of the last 30 years, we have learned that this brain plasticity operates at the cellular level to actually change the wiring of the circuits that control and coordinate our behavior. That is, genetics-- where's my pointer here-- dictates particular source neuron sending long processes to their targets within the central nervous system. That's a product directly of genes.

But the genes don't specify the detail point-to-point ordering of the synaptic inputs of these cells onto their target cells. That ordering is a process of the impulse patterns of activity that the young neurons carry.

Now in many instances, this occurs naturally and normally, and we don't even know it's happening. Because some of this activity can be spontaneous within the source population of neurons themselves. But the circuits become ordered. They become refined.

You go from, for example, a low fidelity map of your retinal surface to a high fidelity map of your retinal surface as a result of the activity that your retinal neurons carry into the brain.

We now understand in this era of molecular cloning, many of the molecules that are involved in this process. That is, we understand them in terms of their changes. We can talk about molecules in the early brain, and molecules in the late brain. And we give them the same name. But they're not the same molecules.

Many of these molecules are made up of multiple gene products. We call them subunits. They are complexes of proteins. They sit on the membranes of our neurons. And yet, their subunit composition and many of the modifications that occur to them change over the course of development.

So they have the same label. But they're not doing exactly the same thing. And so we understand this in a very abstract way.

And so the challenge for those of us interested in the brain is to understand how this is associated with this, and ultimately changes our behavior. I'm going to give you a few examples drawn from the work of neurobiologists across the country to just show you what kinds of changes we know happen at the cellular level.

What you're looking at here is an image of a living brain in a monkey. Actually, this can now be done, and is being done, for patients with epilepsy in humans.

These are sophisticated techniques for looking at detailed changes in the reflectance of neural tissue when it's active-- when it's firing a lot of action potentials, a lot of electrical potentials. The reflectance of the tissue changes. And so if you have a very sensitive photo receptor array and computers that can reconstruct that pattern, you can get an image of activity patterns across the surface of the brain.

And this is a small bit of a surface of a monkey brain. The monkey has a little window secured into his skull. So he's sitting there. He can actually move around with the photoreceptor array above him.

These are the blood vessels. And what this monkey is being exposed to is a pattern of light through only one eye. He's wearing a patch. So what you're looking at is the activity in his visual cortex, projected onto the surface of his visual cortex, from stimulating just his left eye.

These dark stripes are the stripes of inputs from just his left eye. If you took that patch off and put it on the other eye, then you'd see these stripes get dark. You're looking at what we call the ocular dominance columns in his cortex. We have them. All binocular mammals have them.

It is a way in which the brain organizes the initial input from the retina. And it has to keep them very organized, because at the next level of processing, it takes some of this activity and some of this activity and it projects it to the cells that are truly binocular.

Well, optometrists and ophthalmologists have known for decades that if somebody is born with a cataract so that they don't have good vision in both eyes, that in fact just a few months of experience with that cataract, with not operating on that cataract, can result in essentially no vision in the one eye. And neuroscientists have mimicked that condition by raising monkeys for four weeks or so is all it takes-- with a diffusing prism on one eye-- a contact lens that diffuses the image.

If you take that lens off and then image what's going on in this monkey's visual cortex, you find that this disruption-- this precise patterning of activity of the inputs-- has been severely disrupted, severely abnormal. And it doesn't recover. All right.

So this animal, exposed to this early in life, will never recover completely normal vision. If you did that to an adult monkey, no difference. Two weeks. Two months. Six months. You take the lens off, and they're fine.

So the early brain is responsive to the environment. And abnormal patterns of activity can become locked into that early brain.

Another example, and one in which the range of adaptive human behavior is really exemplified, is then documented in native speakers of American Sign Language-- people born deaf who grow up using American Sign Language as their native language. This is a highly visual, very motion-sensitive language in which emotion and meaning are conveyed not only by signs, but by the motion of signs.

We've known for a long time that these people are exquisitely sensitive to motion, much more so than those of us who have normal hearing. These are functional magnetic resonance images of the brain of a hearing person-- an adult hearing person-- and an individual who learned American Sign Language as a child.

These people are in a scanner. And they are being stimulated with a moving visual stimulus. What you see-- this is actually the same brain. Here, we're looking at it. This is the front. This is the back. This is the visual cortex. And now you're looking down at it from the top. Again, the visual cortex.

In the hearing-- the normal person-- as would be true in our brains, all of those who are true hearing individuals, that visual stimulus activates this visual region of our brain. But in the native speaker of American Sign Language, that same stimulus now produces activation in wide regions of the cerebral cortex. Not only the visual regions, but auditory regions, and some regions of the frontal-- the association cortex.

This brain is permanently changed. But if an individual learns American Sign Language after four years of age, you don't see this pattern. Okay. These are changes that our brains are capable of early in life. And after that, that potential is lost. Very adaptive changes.

All right. Well, how do you study this? How do you really understand what's going on mechanistically?

And this is where the experimental approach that utilizes animal behavior, physiology, and whenever possible, molecular biology, comes into play. And I want to tell you about one particular set of experiments because they illustrate the range of these techniques. This is a barn owl. This is work done by a good friend of mine, Eric Knudsen, who happens to be at one of those small universities on the West Coast.

But anyway. Barn owls hunt at night. And they hunt for small prey objects with incredible exquisite localization. And they can use vision. So if they can see it, they'll use vision. But they can also hunt in complete darkness. So they can localize prey objects exquisitely with both the auditory and visual modality.

And this happens to be Roger the owl, and with a time lapse showing how he hones in on this little mouse. Now what Eric and his colleagues have done is use this behavior of the owl to understand how the brain coordinates this visual and auditory information to orchestrate a precise motor pattern, a precise behavior.

And the experiment is to take young owls and train them to hunt for mice, or actually to dive at a point on the floor for which they get a reward.

And you can make small sounds at that point. And in a normal owl, he can do that behavior perfectly well, whether it's a visual stimulus or an auditory stimulus that he's homing in on. That reflects registration of the maps of auditory space and visual space within the owl's brain, and in fact, within our brain. We can do it. We're not as good at it as owls. But we can do something very similar.

Eric then fitted these little young owls-- again-- a major point here, these have to be young owls-- with prisms that deflected the visual image. So now what he has done is he has dislocated the visual image from the auditory stimulus. And at first, the young owl doesn't do a very good job at localizing things. He's getting disparate stimuli.

But if you allow him to fly around in these big aviaries, he will eventually get very good at localizing prey through either modality. And you look in his brain. And what you have discovered is that the auditory and the visual map have come into register. What has happened is that the neural mechanisms of plasticity have brought the auditory map into register with the displaced visual map of space. Okay.

Now you can take off the prism. Okay. A young owl can go back and behave quite normally. You can measure in this experimental animal how good that animal is at adjusting by looking at how displaced his behavior is when he's got the prism on.

And what you discover-- well. We can make you displace here. Thank you. That's a little bit easier. Okay. Thank you.

This is a measure of the behavioral shift that the owl is capable of doing with the prisms on. And you discover that young owls can shift their behavior quite dramatically with the prisms. So the ability to shift is very high.

However, a young owl, given-- and this is only about a week and a half of experience that the animal needs. So you place the prisms on these owls at different times during their juvenile period. And in fact, you discover that as you place them on older and older owls, the ability to shift, to adjust, has gone down. Okay. So that an adult owl can wear prisms for a year or two. And providing he doesn't hurt himself by getting confused in his aviary, you can test him. And he has never adjusted.

All right. Well, what's going on at the cellular level? Actually, there's remarkable agreement-- Does that mean something I should worry about?

HECHT: No. That's just somebody's pager.

PATON: I see. It's not a strange fire alarm or something. All right.

The rule is a very simple one. And it's one that almost everybody in this field agrees on. And that is that simply each young neuron in our central nervous system, in our brain, follows a very simple rule during the course of its development. It has the opportunity to accept input from a variety of different source neurons. And it retains only those inputs that are effective in firing it, in driving it, in activating it.

And there's a very simple phrase that is remarkably easy to remember. Neurons that fire together-- this, this, and this-- wire together. Whereas those guys that don't fire together get lost during development. Young projections are remarkably exuberant. There are many more connections that are made in the early brain than are ever retained in the adult.

And the rule for which ones are retained are that you retain the ones that are effective. And the patterns of activity that they carry determine who is going to be effective, because if this guy and this guy and this guy are active together, their drive will sum on the target and fire that neuron. And there is a mechanism that many of us are actively at work on that feeds back onto these inputs and stabilizes them into maturity.

Now the various maps of our sensory peripheral fields refine during our development using this rule. The inputs are initially disorganized. We know that neighbors within the source population of neurons tend in a probabilistic sense to fire together. So that of the enormous number of connections they can make, the ones that they actually continue to retain and do make are those that come from source neurons that tend to be active together. They refine into a relatively nice point-to-point projection because they use this pattern of activity. Temporal pattern of activity. A very dynamic property.

This is actually an example that I pulled out of an old file, because I think it illustrates how this happens without us being consciously aware of it.

A number of years ago, a student of mine who's now actually a professor at NYU, performed an experiment in which he raised young mice with a click. Click, click, click.

If you have a good amplifier, what it does is it drives all the neurons within quite a broad frequency range coming from your ear together. So instead of when you're just sitting listening say just to white noise, it sounds like nothing to you. But in fact, what your central nervous system is seeing is a lot of different frequencies all together.

It's like white light. White light is all different electromagnetic frequencies together. Noise is frequencies, auditory frequencies, played together.

Now during normal development in mice, if you look at the sensitivity of cells in the central nervous system, where they have what we call very broad tuning curves-- that is, all neurons will have a frequency to which they are sensitive at very low intensities. But then as you get further away from that frequency, the cells are much, much less likely to respond to a low intensity.

And so this is a tuning curve. Adult organisms of all vertebrate species, or the vast majority of vertebrate species, have very narrowly tuned tuning curves. But young organisms have broad tuning curves.

When Dan raised these mice in this click environment, he discovered that this normal process of going from very broad tuning curves to very narrow tuning curves had basically been stalled in the range in which the click was giving equal input across the frequency. So the auditory nerves were now not giving information about who was whose nearest neighbor.

Instead, over a broad swatch of neurons, they were firing together. And the process had been jammed. So the cells didn't lose frequencies, inputs from frequencies that were quite disparate. Instead, they retained many of them. And in fact, these animals have very poor frequency discrimination. Okay.

So the bottom line is, the reason the Bach family might all be great composers may be partly genetics. But partly because early on, they heard a lot of very nice music.

All right. So who are the molecular players in this game. The molecular players are the molecules that mediate the communication between input neurons and their targets. These are the neurotransmitters and the neurotransmitter receptors.

Neurotransmitters are very simple molecules. And in fact, the major neurotransmitter in our brain is the very simple molecule glutamate. But there are close to now 30 different defined glutamate receptors on our neurons. And we now know that one particular category of glutamate receptor is very, very important for the early development and early operation of this plasticity mechanism.

I thought I'd show you a picture of a real neuron instead of a cartoon. This is one growing in culture. And all these little dots are fluorescently tagged glutamate receptors. So there are a lot of them on our neurons.

The one I want to call your attention to is one that is known mnemonically as the NMDA receptor-- the N-methyl-D-aspartate receptor. It's called that because N-methyl-D-aspartate is an exogenous molecule that happens to bind with very high affinity to this receptor. And it's very useful pharmacologically for studying this receptor.

But it's unusual in that this receptor is very, very exquisitely tuned to respond when there's a lot of input activity coming in simultaneously to a neuronal membrane. All right. And it lets in a lot of very important ion calcium.

And we don't begin to understand all of the biochemical cascade that this calcium mediates. But we do understand that this molecule, if it's disrupted in the early brain, disrupts many of the refinement processes that I've been telling you about. And this is an example from a study that we and a set of collaborators did a number of years ago.

This is a rat brain. And this is a midbrain visual region of a young rat. And what we've done here is labeled the population of processes that come from just the posterior pole of this young rat's retina. These processes are labeled with a fluorescent dye. And you can see that on postnatal day four, they go all throughout this region. They terminate throughout this region. By day six they are beginning to show a very specific region of projection, although there are a lot of processes in abnormal regions. By day 12, they're restricted to one region of this visual structure in the central nervous system that is very characteristic of the adult.

Now this is happening on all points of the retina. So very early in life this visual map is very diffuse, very disorganized. And it refines.

We can go in and block this N-methyl-D-aspartate subtype of glutamate receptor, actually block it with a polymer that Bob Langer developed, and you'll be hearing about shortly. You can put in this polymer and infuse it with an antagonist to this receptor. And when you do that, you discover that these young rats, which can't use their an NMDA receptor, retain a lot of these exuberant projections. So the refinement process is blocked when this receptor is blocked.

I'm not going to go into a lot of details. And I'm going to stop very shortly. But what I want to emphasize is that we now know that simply blocking this receptor not only blocks the development and refinement of many sensory maps, but it blocks the realignment of auditory and visual maps in the ow. And it blocks many of the activity-dependent changes in visual cortex.

This is a very important molecule. It's certainly not the only player in this game. But it's an important molecule. So what can we do with this information? How does it help us?

To understand the process and hopefully to manipulate it for the better of humanity. Well, in mature brains, the function of this receptor is low. And in fact, it's low and lowest in those regions where the potential for adaptive neural plasticity is low. All right.

And in fact, it changes its sensitivity quite abruptly in many regions of the brain. And one can then say, well, can we understand what changes that sensitivity? Can we reactivate developmental plasticity in a relatively mature brain, and allow for recovery from early deprivation, or many instances of CNS trauma where you might want to recover normal wiring, or in fact help children and young adults recover from early neurological dysfunction?

Well, we're now learning a lot about the changes that happen in this receptor. We know that there are mature type subunits that get added to this receptor as the brain matures. You take out baby subunits. You activate new genes. You plug their products into the receptor. And that changes the function of the receptor.

We also know that there are modifications that occurred at discrete times in development to this receptor, sitting on already established neurons. And we're learning what those molecules are. And we're learning how to manipulate their function. So indeed, there is a great potential for designing drugs that can target those enzymes that alter the function of the receptor.

And we can also reintroduce the baby genes into regions of the mature brain. And those will allow us to reactivate the function of this molecule in a mature brain.

But there's a catch here. All right? We have to know what brain region to target to. Okay.

You can't just reactivate this receptor throughout the brain, because indeed, if you overactivate the receptor, you can cause brain cell death. All right. So we have to know what regions we need to reactivate it in. And we have to know the cascades within the cell that activate the cell death process. I mean, this receptor is also a major player in the death of cells after brain trauma. So you would like to turn it off when you don't need it. You want to turn it down.

Okay. So we can answer those questions maybe in 20, 30, or 40 years. Okay. But I want to add one last thing, and then I'll get down from this pulpit.

These are data from the owls again. Okay. And what this data is showing you are the results of experiments in mature owls. All right? So Eric takes these animals, raises them. Some have prism experience early. Some are raised as just normal owls.

And he discovers that an owl that has never experienced a prism can never readjust. Okay? Never readjust its behavior. However, the owls that had maybe just two or three weeks of prism experience as juveniles quite rapidly learn to cope with the prisms again. You have introduced into these owls' nervous system a plasticity that isn't normally there by giving it a broad, wide experience as a juvenile.

And I think quite apart from molecular biology and biotechnology, there's a lesson here for educators and sociologists as well. Broad experience in early childhood when the brain is plastic will provide a potential in adulthood for tremendous plasticity that I think many of us have just never tapped. Thanks.

[APPLAUSE]

It's nice to see you.

HECHT: Our next speaker will be Bob Langer. And Bob is going to shift. Bob is an engineer, some would say a consummate engineer. And I've had the privilege of hearing Bob several times this year. And I'm always amazed. Robert.

[APPLAUSE]

LANGER: Thank you very much. It's a pleasure to be able to speak to you today. I guess we're going to shift from really good biology to now engineering. And my background is that of a chemical engineer.

And what I'd like to discuss with you is the area of biomaterials. And I'd like to give you a talk at least to give you a little bit of my view of the future, of how biomaterials may affect our lives. I'd like to focus in on two ways that will maybe see this.

One is ways of delivering drugs better. And the other is the idea of creating brand new tissues in the human body. So let me start with an example of what's already happening, and then take you through some of the work we've been doing in both of these areas.

This is an example of a polymer-based system to deliver a drug. It's a nitroglycerin patch. People may have seen these. People may even use these.

But this field of what we call controlled drug delivery, which involves putting drugs in plastics and delivering them over long times at steady rates is a very new field. If you just went 20 years ago, there would be none of these on the market.

Today we see this to be a thriving field. Last year, for example, in the United States, there were $16 billion of sales of these kinds of systems. Not just this one. But all different types.

And it's not only that you can just prolong delivery. You can tremendously reduce the side effects associated with conventional drug delivery. Why is that important?

Well, let me just give you a couple statistics. Last year in the United States, over 100,000 people died taking prescription drugs exactly the way they were supposed to. That's four times the number of deaths caused by AIDS in this country.

Last year, if you look at the health care costs associated with people taking drugs exactly the way they're supposed to with mishaps, it's $136 billion dollars. That's more than heart disease in the United States. So these are huge problems.

If you could deliver the drugs at steady rates rather than the peaks and valleys which you normally get when you take drugs the conventional way-- the peaks potentially causing toxicity, the valleys basically being ineffective-- there's the potential to eliminate many side effects caused by conventional drug therapy. This is just one example which can prevent problems associated with liver damage and so forth.

But not only can you use technologies like this to deliver drugs for one day, you can use technologies like this. This is a little implant placed underneath the skin called the Norplant. It's incredibly powerful technology. And it can deliver a drug with a single implant for over 2,000 days, or five years.

This is a system, as I said, called the Norplant. Not sure what the background does.

Anyway, let me tell you a little bit about some of things that we've been trying to do with that as a little bit of background. The molecules that are being delivered here-- and this is also very new system first approved by the FDA in 1991. They're for very, very tiny molecules.

What I became interested in going back to the early 1970s is, could you deliver large molecules? Polypeptides, peptides, and so forth. These molecules are incredibly difficult to deliver. They're very hard to swallow because they get destroyed. They don't get absorbed. They're very hard to take by these patches for the same reason.

If you inject them, you're often faced with very short lifetimes in the human body. So if you want to use these on a chronic basis, it would be critical to have a way to deliver them in an unaltered form, and yet protect them from harm.

But when we started our research, is the projector-- Wondering if somebody could help me. This doesn't seem to be working too well. Thank you.

So when we started our research, the conventional wisdom in the polymer field was, you would not be able to use these kind of approaches, to use polymers to slowly deliver big molecules. If you look at the literature in the 1970s, you often see quotes like this. This is a direct quote. I'll just read you this one. Says "The agent to be released is a small molecule with the molecular weight no larger than a few hundred. One would not expect that macromolecules, for example, proteins, could be released because of their extremely small permeation rates through polymers."

Actually, this quote was very helpful to us in getting a number of patents in this field. And it was against this background that we started this work. What did people think?

Well, what people thought at this time is that if you had a big molecule, and you're trying to move through a polymer, it's kind of like diffusing through a brick wall. Alternatively, you might go through like something like Swiss cheese-- you know, with big pores. But that's going to be really fast.

So the question was, how could we get release, but do it slowly? So we spent several years-- I don't think this is working. Maybe somebody could control the slides if I just say next slide.

So we spent several years, we're able to make little microspheres like this. Here's one cut in half. Next slide.

And if you did a release curve, what we'd see is that you could take different proteins-- this is 14,000, this is a quarter of a million, these are in between-- you could get release for over 100 days. And using various engineering approaches-- next slide-- you could get constant release for really any rate of time that you wanted.

Now the question is, how does this work? And I'll just tell you a little story. You know, when I started my work I was a postdoc at the Children's Hospital in Boston. And I'd never given a talk at a meeting before. So in 1976 I was asked to give the very first talk that I was going to give, This was at a big polymer meeting in Midland, Michigan.

Midland, as you may know, is where Dow and Dow Corning is. And big polymer center. So I stopped work about three weeks in advance of this talk. And I kept practicing my talk into a tape recorder. This was before VCRs were available. And I kept doing it over and over because I was really nervous.

And I was pretty young at the time. And so when I gave the talk and I actually felt I did okay. I didn't forget too much of what I intended to say or anything like that.

So when I get done with the talk, I felt there were all these more elder polymer engineers and chemists, some pretty famous people in the audience. And I was this young guy. And so I thought when I was done and I did okay, that these scientists being nice people, would want to encourage me, this young guy.

And so when I got done, a number of them came up to me and they said, we don't believe anything you just said.

[LAUGHTER]

That was kind of my introduction to how scientists treat other scientists. I keep seeing that, unfortunately. Not from MIT graduates.

So it wasn't until about three years later in 1979 that several groups had repeated what we'd done. And then the question shifted to how could this possibly happen? So what we did-- next slide, next slide-- is we used techniques that are normally used in medicine, pathology. And we cut very tiny thin sections through these polymers.

One of them we used was ethylene vinyl acetate copolymer. And we cut a thin section through it. And here is a 5-micron thin section. There's no drug in it.

And actually, if you did a permeability study, what you'd find is that the molecules were 300 molecular weight or greater, they would not be able to diffuse from one side of this very thin membrane to the other. So how could this happen? How do we get these molecules out?

Well, in the next slide what we see is that if we put a protein in-- and now we put a red protein in, myoglobin-- we get what we call a phase separation. Polymer is white. Protein is red. This is before any release has taken place.

But if we release these systems for, say a year, and then cut a thin section, the next slide shows that left behind after the release are pores. Notice how different this slide is than the one you saw two slides ago. So these pores are made by the way we incorporate the drug.

And what you can see is when you look at them that the pores can be significant in size. But also we've done a lot more microscopy. And what you can see is these pores have very tight constrictions between them, kind of like going trough Logan Tunnel.

And they're incredibly winding and tortuous. In fact, one of the analogies I use when I go different places to explain this to people why this works is I tell a myself being from Massachusetts, you can kind of think of this as like driving a car through Boston. And by controlling the pore structure, some of my students over the years-- and there's a number of engineering approaches we've learned to do this-- by controlling the pore structures, we can take very tiny systems. We can make them last anywhere from a day to actually over three years.

In fact, the biggest problem I've had as a professor in getting really long-term release has actually not been the design of the system. It's actually been my ability to find a graduate student who would do the experiment that long.

And actually, I was very, very fortunate in the late 1980s. I had a terrific graduate student named Mark Saltzman. And Mark actually did this experiment, kept releasing it and releasing it. He did it for three years and two months.

And then one day he came up to me and said, Bob, you know, I've been thinking. He was doing his first PhD thesis. He said, Bob, I've been thinking. What will I do if this experiment now fails? He said, do I have to repeat it?

[LAUGHTER]

And so we stopped at three years, two months, and five days. But I would like to think with the right graduate student, we could go five, six, seven years. We haven't found him yet, though.

But anyhow, Mark's actually doing very well. He graduated. He's now full professor at Cornell. So there's good endings to these stories.

At any rate, what could you use these for? Well, one type of thing you could use these for is really in what we call bioassays for the type of very elegant studies you just heard in the last talk. And that's been one good example. And there are a variety of different bioassays for delivering what we call information on molecules for just understanding things in biology better.

But in addition to that, there's potential clinical applications. The next slide shows you one that Larry Brown, one of our students, worked on. And that was taking an aspirin-sized pellet and using it to slowly deliver insulin-- this is still in animal trials-- for over 100 days with a single little implant. And these actually have been used clinically, these principles.

The next slide shows you an example. We had a group of Japanese scientists coming every year to our lab in the 1980s. And this is at Takeda. And they work with Abbott. And they came up by using these principles to make little micro capsules very much what I just showed you. But they put it in this little bottle like this. This is called a Lupron Depot.

And this system takes a peptide that's about 1,200 molecular weight. It's actually one of about three systems that does this. It's about 1,200 molecular weight.

You put it in these little polymers. It will slowly deliver this drug for four months. You absolutely need to put it in the polymer. If you didn't, it would be destroyed right away when it's injected in the body. And it's so big, you need to inject it. You can't swallow it.

This today is the most widely used way of treating advanced prostate cancer, endometriosis, and precocious puberty. There are a number of other of these on the market. And I think what we're going to see as we move into the next century are many examples of these.

Probably the very next one will be a one-month delivery system developed by Alkemes, a local company, and Genentech, for delivering human growth hormone, which now little children have to take once a day. Pretty soon they'll be taking it once a month. But there are many other examples I believe we'll also see using these kinds of approaches.

But I'd like to think that this is just the tip of the iceberg. Even though I've shown you that we can deliver drugs at steady rates or at decreasing rates, wouldn't it be nice someday in the future to be able to really regulate and control what you do? So a few years ago-- next slide-- we had this idea. Could we actually make smart systems?

And the first way we thought about doing this was with magnets. And our idea was to take an elastic plastic, put little magnetic beads in it-- this is a cartoon, not drawn to scale-- and then put drug in it, which are represented by these little dots.

And our thinking was is that under passive conditions the drug would diffuse out. But when we apply an oscillating magnetic field, the beads kind of squeeze these pores. So a lot more drug comes out. And the idea is almost like milking a cow. I'm sorry. Do you--

It wasn't working too well. So. Anyhow. I just wanted to make sure everybody here was working.

So the idea would be you'd almost do like milking a cow and squeezing more drug out of the pores.

The next slide just shows a picture of what these look like. Here's an example of one. Here's x-ray showing you the magnetic beads.

And you might think, how would you get these to work someday practically? So our idea is we make a little implant, maybe in the wrist area. And then maybe we give somebody a special wristwatch-like device that you could have pre-programmed, or you could turn it on say, at you meal time.

Now we weren't able to do this right away. But I've mentioned this to one of my very good graduate students at the time, Elazer Edelman. This was in the early '80s. Elazer is now a full professor at MIT also.

And I said to him, could you try to make a rotating system. Simulate a wrist watch. But of course, we weren't able to miniaturize it right away. But what I wanted is something that would rotate around to create an oscillating magnetic field.

The problem was, this was very early in my career. And I didn't have much research money. And Elazer was a very ambitious student. And so I said, Elazer, could you try to make something that would rotate around to test this out? But I said, could you do it for less than $50?

[LAUGHTER]

So next slide. So what he did, he went home, and he actually had a record player-- a Victrola-- that he'd had since he was a small boy. He took it apart. And he made this system.

Here's the record player motor. He actually made, this by the way, for $49.50. So here's the record player motor. He attached it to two Plexiglas disks. The top is stationary and has vials. And the bottom one will actually have magnets on it. Here you can see one. And it will rotate around to create this oscillating field. And it rotates like a 33 or 45 or 78 RPM.

The next slide shows you what it looks like in the lab. Vials. Magnet. And what you see on the next slide-- next slide-- is that when you turn on the oscillating magnetic field, you can get release rates up to 30 times what you get when it's off. And the height of these peaks can be controlled by the strength and the frequency of the magnetic field.

Now you always want to test these things at least in animals, if not people. So in animals what we wanted to do is again take diabetes as an example. The next slide shows what we did. We took a little system here. This has about a two-year supply of insulin for a rat.

To give you an idea of small it is, we put it next to a ping pong ball and a ruler. And then what we did is we placed this underneath the skin of rats. But we had to change our triggering device a little bit for these rats.

Next slide. Here you see it. And you can see that these rats are really very, very happy. This is the treated group. Controls, I won't tell you about.

But at any rate, we give them 20 minutes treatment to the field. By the way, sometimes people think when they see this that we're rotating the rats. But they're not. Remember, the rats are in the top part. That's stationary. It's the bottom part that rotates around.

So we give them 20 minutes treatment with the oscillating field. The next slide shows you the results. And diabetic levels about 400. This line is the time of implantation.

The blue bars show you passive delivery of insulin. The red bars show you what you get just 20 minutes later. And each time you get a significant lowering.

So this shows you you get a response in vivo in 20 minutes. We're still not doing it on people yet. But we hope at some point to be able to do that.

But we have, as have other people, worked on new ideas, and hopefully even better ideas for getting smart systems. I'd like to tell you the latest. And the latest idea we had is a few years ago I had this idea. I was watching a TV show that talked about Intel-- you know, the company that makes chips.

And I thought, most people when they look about things on chips, they think of computers and stuff. I always tend to think about drug delivery, because that's what we do.

So I thought, gee, wouldn't it be interesting if we could turn chips into drug delivery systems. So the next slide shows you our idea. And I did this. I called up Michael Cima, who's a professor in material science. And I got John Santini, a terrific graduate student, to work on this.

And our idea is, what if we took a silicon wafer-- this is just a prototype, by the way-- and we build little wells in it. And we could basically build in a little wafer, here we just did three. But we could actually do 1,000. It's actually very easy to do.

And the idea is actually kind of simple. On one side of the chip we have an impermeable epoxy so the drug can't come out. On the other side we put gold. Could be other things. But gold we found is quite good. It's relatively safe.

And the idea is that if we simply apply 1 volt selectively to this wire, we can dissolve this cap-- this gold cap. As soon as we do, our theory is all the drug should come out.

And basically this idea is that you could essentially have a pharmacy on a chip, that you could but in multiple drugs. And it could be delivered multiple times a day. If you want to someday you could put in microprocessors or biosensors so you could get remote control or smart systems.

Well, how do we go about doing this? Well, the next slide shows you actually an example. We've made these. They're actually the size of a dime. Here's a dime. And here's the chip.

What you're looking at are both sides of it. On one side you see each of these little dots. And again, we could make these dots any size. Here we've made them small.

But each dot could contain a different drug, or the same drug in a different dose. Here's the other side where you can see the wells. So you have anodes, cathodes.

And basically what we can do is selectively trigger any one of these wells. And when we do, the drug inside it should come right out. Let me show you that. Let's do a scanning electron micrograph of just one well over time.

The next slide shows you what one of these wells look like. Scanning electron micrograph of one gold cover. But let's look at it as we apply just 1 volt selectively to this one. And look at it over what happens in 10 seconds. The next slide shows you that.

What you see is as soon as you apply this volt, 1 voltage, the gold cover comes off. And as soon as it does, the drug inside-- if there's no resistance, like it's more like a solution-- could come right out. And to illustrate that, let's do a release study as shown on the next slide, where what you see are one drug now coming out multiple times. This is at 1.5, 2.5, 5.5 days. This is all arbitrary.

We could make them come out whatever time we want by simply triggering those individual wells. Here we've put simply one drug on. But what if we wanted to do multiple ones?

Well, the next slide shows that. Where we've put two drugs on. And again, we've done remote-- by applying 1 volt, we make it one come out at 25 hours, another 35, another 50, another 60. In the long run, what we hope to achieve with this is as I mentioned earlier, you might have systems where you could implant and do remote control like telemetry, like you'd open up a garage door to regulate this. We're trying to work on ways of doing that now.

You might even incorporate biosensors on to these to make a smart system with feedback control. There may be even systems-- we're working on ones that you could swallow, where you wouldn't have to do remote control, where people might have to take multiple drugs multiple times of day, and they would just come out when you wanted them to. Again, this is for the future, not yet for today.

Now so far what I've done in this talk is I've told you about what we could do with existing materials and make them better, and do things you could never do before. But one of my other concerns has been the materials themselves. You know, I'm a chemical engineer. But I worked in Children's Hospital in the 1970s for a few years. And one of the things I began to observe was how do polymers find their way into medicine?

Being a chemical engineer, I would've liked to have thought that it was people like myself, or material scientists, or chemists that played a role in bringing materials into medicine. But that's rarely been the case. Almost always what's happened is that clinicians, or sometimes companies, have been trying to solve a medical problem. And what they've always done is they've taken an off-the-shelf polymer that kind of resembles the organ or tissue they're trying to fix. And they use it in a human being.

So for example, next slide. This is all true, by the way, what I'm going to tell you. If you look at the history of how polymers have found their way into medicine, what have we seen?

Well, let me just give you a couple examples. Back in 1967 at the National Institutes of Health, they were trying to design an artificial heart. This is actually the origin of the artificial heart. And they were trying to figure out what material they wanted.

And so the clinicians there said, well, we want something with a good flex life. And they thought about this and said, what object has a good flex life? They said, a ladies' girdle. And they said, what's it made out of? Its polyether urethane.

So they decided to make the artificial heart out of the polyether urethane. So what's happened now that we're 32 years later? Well, actually, the mechanical engineers, they've done a pretty good job. The heart beats long enough.

But if you read various journals like even Time or Newsweek, what do we see? The reason the artificial heart hasn't worked very well is that very often, when the blood hits the surface of the artificial heart, it forms a clot. And the patient gets a stroke. And they may die.

And yet to me it didn't seem that surprising that something was designed to be a lady's girdle might not be the best thing to put in the human body. And that's still what's used.

This problem, it pervades all of medicine. So dialysis tubing was originally, and still is, sausage casing. Vascular graph-- that's artificial blood vessel-- was a surgeon in Texas going to a clothes store, finding what he could sew well with.

And I don't want to minimize these approaches, because they've been able to progress to be made. But there are also problems. As you know, you can't make small diameter blood vessels.

Breast implants-- actually, I hope there's no lawyers here. But if you look at breast implants, the origins of these-- one was a lubricant. And the other actually was in mattress stuffing. It's true. I mean, you can kind of think of the logic that they used.

[LAUGHTER]

But I mean, it is true. I realize it's funny. But it actually is true. I mean, that's exactly what's happened.

So what I started thinking about was, well, maybe we can use a better approach. What if we take rather than off-the-shelf materials, what if we ask the question-- what do we really want in a biomaterial from an engineering standpoint, from a biology standpoint, from a chemistry standpoint? And then could we synthesize them from first principles?

So we picked an example. And that example is biodegradable polymers used in drug delivery. So the next slide shows you what we were thinking about.

Most polymers that are used in drug delivery dissolve like this-- show bulk erosion-- start out like this, become spongy and fall apart. And that can lead to bursts of drug release, which if you had a very toxic drug like insulin or an anti-cancer drug, could be fatal.

So from an engineering standpoint, what would you want? Next slide. So we said from an engineering standpoint, you'd much rather have this. That's the way like a bar of soap dissolves. And that would prevent any possibility of dumping the drug.

And so without going through all the engineering design analysis that we did-- and we spent a lot of time on this-- we came up with a polymer that we thought would solve this problem. Next slide.

And this is the one chemical structure I'll show you. So you might think here's a very hydrophobic unit, one that repels water. Here's another unit. But just think about it that we've got two units-- A and B. And what I'll show you on the next slide is that by simply regulating the ratio of that A to that B, we can make these last anywhere from like about 8% dissolves in 14 weeks if you use 0% of one of them. Whereas all of it will be gone in two weeks if you use 79%.

And so you can simply by dialing in the ratio of the A to B unit, make these last from almost any length of time you want. So if you could do that, you have tremendous control. And you could begin thinking about applying these to different diseases.

So let me give you an example. Next slide. About 12 years ago, Henry Brem, a neurosurgeon, came to see me trying to feel find a better way to treat cancer, in particular a very bad form of cancer-- brain cancer-- in its worst form-- glioblastoma multiforme which is uniformly fatal. And regardless of how it's treated, the mean lifespan is less than a year.

Now the drug-- in addition to these terrible statistics-- the drug you use for brain cancer-- next slide-- BCNU-- is a terrible drug causing terrible side effects when you give it. Normally it's given intravenously. But it travels throughout the whole body, causes terrible side effects to like the liver, kidney, and spleen.

So our idea is shown in the next slide. Local therapy. Could we do this? Could only allow the neurosurgeon like Dr. Brem to operate on the patient, but line the surgical cavity before he closes the patient with a polymer containing BCNU?

Now BCNU normally has a lifetime in the body of 12 minutes. But if you put it in the polymer, it's protected. It will last as long as a polymer lasts.

So what we wanted was a degradable polymer that would not accumulate in the brain, a polymer the dissolved by the surface erosion so it couldn't dump this really toxic drug. And finally from animal studies they'd done at Hopkins, our collaborators, we wanted it to last for a month. And you could easily do that by adjusting those monomer ratios.

Now, the polymer protects the BCNU from degradation. But the most important thing is local therapy. High concentrations in the brain where you want it to be, and almost no concentration in the rest of the body where it caused these terrible side effects.

So I thought I would also tell you how were these ideas greeted by the scientific community. So the next slide-- I won't go through this in detail. But you know, when you're a professor, as all the people speaking today are, generally you have to write grants, usually in all of our cases this morning, probably to National Institutes of Health. And when you write these grants, they write back comments about them from other scientists.

So I decided I'd show you some of them. And I entitled this slide, This Approach Will Not Work Because. I simply pulled this out of different grants. Without going into detail, every two years we got a different objection. Each time we would solve it, we'd get a different objection.

Probably all my colleagues would have similar stories. It's actually the way, again, scientists work. But the point is that every two years you get a different one.

And finally by 1996 we'd taken this from a concept we drew on paper to a brand new treatment that was actually approved by the FDA. And actually, the only treatment, as I will show you in a few minutes, that was approved by the FDA for treating brain cancer in the last 25 years.

But this is basically sort of what happens. I'd like to actually show you a picture or two of how this looks. But if anybody is squeamish-- and I mean this seriously-- doesn't like the sight of blood, don't look at the next slide or two. But let me just show you on the next slide what the operation looks like.

You put little wafers the size of a dime in the human brain. Next slide. There's about seven or eight of them. Next slide. And then you close up the brain.

I always show those slides very quickly. And the reason I do-- it's very hard to get good advice on your talks. But a couple years ago, my wife, who's a neuroscientist-- got her PhD with Ann Graybiel here-- came to one of my talks. And I asked her at the end-- this was a talk to chemical engineers. And I asked her at the end of the talk, I said, well, what did you think of the talk?

She said, well, it was okay. That was actually pretty good praise from her. But she said it was okay. But she said, that 15-minute period where you had those two slides on and you explained the whole operation, she said, I don't know if you noticed, but all those chemical engineers were turning green and looking at the floor.

So I've always shown those slides very quickly after that, except that I want to tell you a little bit of a sequel to that, which was that a couple years ago I was asked to give a speech actually at what's called the Princeton Conference, where a lot of neurologists and neurosurgeons go. This was actually also a dinner speech.

And I showed those same two slides. And I went through them very quickly. And at the end of the talk actually, a number of the neurosurgeons came up to me. And they said, you know those two bloody slides you showed, you could've kept them on as long as you wanted. But they said, those chemical formulas!

[LAUGHTER]

So I have learned that depending on what audience I speak to, I keep different slides on different lengths of time. To show you a little bit of the clinical data, this is from what's called a phase three trial, final stage clinical trial. This was in Europe.

And what you see is that at the end of the year, in the treated group, 63% of the patients are alive. At the end, the control group, which are best ways of treating brain tumor at the time, 19% alive. At the end of two years, 31% in the treated group. 6% in the control group.

And again, our hope is this is just the tip of the iceberg, that this may, at least for tumors that are relatively localized, you might be able to have ways of doing local therapy. And in a broader sense, it makes maybe people think of more rational ways of designing polymers for many, many diseases.

I'd like to spend now the last final minutes on not just the use of materials for treating drugs, but for actually delivering human cells to the body. Let me motivate this with another picture of a patient.

This child-- next slide-- who has liver failure. And so he's bloated like that because the liver with all the edema. Next slide. Now, the only way a child like that can live is someone else has to die. And you give them a transplant.

But the problem with that-- just take liver-- is last year in the United States, 30,000 patients died of liver failure. There are only enough donors for 3,000 patients. So many people, including that child, unfortunately, died. Just can't get a donor. How do you solve that problem?

Well, the way Jay Vacanti, who's my collaborator in a lot of these studies-- Jay's a transplant surgeon at Mass General Hospital. We came up with this idea about 16 years ago that maybe what you could do is take-- well, maybe I should say it this way. If you take isolated dissociated cells and inject them into the body, not much happens.

If you take those same cells, put them close enough together, they're actually smart. They can actually reform structures. And a group, for example, at Berkeley has shown that you can take mammary epithelial cells, put them close enough together, they can actually form acini and make milk.

So our thinking was, could we make polymer fibers where we could take any cell type, line them up on these polymer fibers, grow them in vitro-- and again, a lot of this involves the right kind of polymer synthesis and design-- but grow them up in vitro and form a new tissue? This is our strategy. It allowed them to recreate their structure outside the body.

So let me give you a couple of examples. The next slide just shows you a picture of what the polymer scaffolds look like with liver cells. So we have very, very-- we have to design these so they have a very high surface area to volume ratio so you could put a lot of cells on. So I'll give you three examples in terms of closing the talk.

First one is cartilage. There's over a million operations each year where people need new cartilage. Our idea is someday what you'll be able to do is go in with a minimally invasive procedure in arthroscopy, like say go into the ear-- we're actually even doing this on patients now. But your ear is all cartilage.

Take a few cells, multiply them outside the body, which you can do. That'll solve the donor shortage problem. Add on a scaffold that you could make in any shape you want. So let's say you wanted to do plastic surgery on a person someday, make a new tissue. Well I'll show you an example on a nude mouse.

Next slide. Here what we decided is we want to redo this guy's skull. We made a scaffold like this, put cartilage in it, and redid it.

On the next slide what I'll show you is another nude mouse. We re-did his cheek. The next slide shows you if you open these animals up, what the tissue looks like. And it looks like cartilage, acts like cartilage, and so forth.

But let me get to more practical applications. There are children who don't have ears. And Joe Upton, one of our collaborators, wanted to make new ears for patients.

So Linda Griffith-- one of my former postdocs, again now a professor at MIT-- next slide-- made a scaffold in the form of a human ear. Here is the skin electron micrograph of it where you see the cell matrix. And Chuck Vacanti, another collaborator, actually placed this on a rabbit.

Next slide. And you see the human ear on the rabbit. Now next slide. Now you might say, well, why do you do studies on animals? The reason we do studies on animals is because we want to find out if it's safe to do them on people.

And here's a 12-year-old boy. This 12-year-old boy, though, unlike other 12-year-old boys, doesn't have a chest covering his heart. And he likes to play baseball, like other 12-year-old boys.

But if he were hit with a baseball in the chest, he could die, because he has no protection. So what we did is we fashioned the scaffold using these polymer principles, and Dr. Vacanti, my collaborator, put a new chest in him. The next slide shows the image of that, showing his chest being protected.

He's now 16 years old. In fact, Business Week had featured him last year riding a skateboard. And he's doing very well. So that's just an example.

Our hope is to do many more of these as we move to the future. Next slide. So that's one example, which we're even doing today-- beginning to do today on patients.

But what about the future? Well, one area could be nerve regeneration. There are actually polymers that we thought could be very interesting, because they conduct electricity. It's been shown that electrical charges can stimulate neuronal processes, and even stimulate regeneration of transected sciatic nerve ends.

And we have found that an electrically conducting polymer known as polypyrrole can have profound effects on stimulating nerve regrowth, at least in vitro. And we're beginning to see that in vivo.

And so what we did-- this was work done by Prasad Shastri and Christine Schmidt-- next slide-- is to take a sciatic nerve model, make a gap, and do what we try to call like a nerve guide. Like, say you severed your finger. Could you have a guide where you put the nerve endings from one side into the other, and have them grow across the distance?

And that's basically what was done here. But one of the key things to look at-- next slide, next slide-- are the density of nerve fibers. And here each of these little circles is a nerve fiber that we see on the polypyrrole six weeks later. To compare that to regular nerve-- next slide-- looks quite similar.

And yet if you do a control polymer, even some of the ones the people have explored as a nerve guide-- next slide-- you don't see nearly the number of nerve endings. This is silicone. Here's polypyrrole. Tyrell So there's a huge difference. Transmission electron microscope shows similar data. And now, of course, the key is to do electrophysiology studies, which we've been in the process of doing with people at Children's Hospital to really look at function.

Again, this is much more to the future then it is today. But I think it's something that may someday be possible, which was kind of what I'm trying to illustrate in the talk. But with that in mind, let me return to the present, and still try to get across I think some of the very exciting things that can be done even with technology that we've already developed.

And the last example that I'd like to give you concerns the fact that we've actually-- as many things at MIT are-- licensed these to companies, and let the companies work on them in different indications. We licensed some of this technology to a company in San Diego. And they've been working on creating new skin for burn victims.

So let me end this talk with that example, show you a picture of a two-year-old boy in the next slide, who's badly burned. Two-year-old child.

But-- next slide-- using the principles I just mentioned, you can make a polymer cell system. This is a human dermal fibroblast from neonates. And what you can do on the next slide is layer that over the child at the time of the burn. And on the next slide, you see him three months later. I'm sorry. Three weeks later.

And on the next and last slide, you see him six months later. Next slide, or slides off. This, again, was approved by the FDA in 1997, and basically again illustrates what might be achieved by using these kinds of approaches.

And it is my hope that as we move to the future that what I've talked to you about today are just but a few examples of the things I hope materials can do in terms of relieving suffering and prolonging life. Thank you very much.

[APPLAUSE]

Thank you.

HECHT: And our final speaker, Dr. David Page.

[APPLAUSE]

PAGE: Good morning. We're going to try to keep some time for questions at the end as I understand. So I'm going to try to be fairly brief here.

I want to share with you some ideas. Just plant in your mind. I know when many of you were here at MIT, biology was not yet a required course. And DNA was not the common subject that it is today. But in any case, I want to introduce you today to the notion of DNA, second concept spelling differences, and the third thing-- a method for finding critical spelling differences.

So that's what I'm going to summarize for you in the next few minutes. Let's see if this works for me.

Could I have the first slide? This is sort of a cruel test. But. If I could have the first slide, please.

Oh well. This is the second aspect of the cruel test. Ah. Can you see it? No.

So now too much light. Okay. Too much ceiling light. Is it visible now? Okay. A little better. Okay.

Well, as I understand it, today brings with it the third leg of the Triple Crown. And so I present you Willie Shoemaker. He's not going to be running today. But he and Wilt Chamberlain agreed to pose for this slide.

The point here is that we human beings come in a great variety of shapes and sizes. How do we come to be so spectacularly variable, not only in our outward appearances, but also in our disease susceptibilities and our state of health? Is it purely bad luck that some of us will get Alzheimer's while others will not? Is it purely bad luck that some have a heart attack at a young age while others do not?

Well, if we could understand the causes of disease at the most fundamental level, if we could understand why some of us are at particular risk, we would have powerful new tools to bring to bear in our search for cures and prevention. And we might even be able to customize our own medical care. That is, we might even know which disease or diseases each of us is most in need of preventing.

Well, in the last 40 years, biology has undergone-- it works! Biology has undergone a transformation nothing short of a revolution. And that's why biology is now a required course here.

We've come to understand that the central player in biology is DNA. And basically this slide summarizes the now required course. But of course, we do take a full semester to present the material. So I won't go into great detail.

But basically we have this central idea that you've got a blueprint. In each of the 10 trillion cells that make up your body, you've got a blueprint this material called DNA. And somehow that information gets translated ultimately into a series of proteins.

This DNA makes RNA makes protein is really the central dogma of the revolution that's occurred in biology. Now it turns out that each of us has 3 billion letters-- 3 billion letters of DNA. Our instruction books contain 3 billion letters.

And it turns out that any two humans are virtually identical in the instruction book. 99.9% identical. So as you look around the room, it is not simply a matter that you are all MIT alumni that makes you similar. In fact, your DNA sequences are 99.9% identical to those sequences of anyone else on the planet.

But there are differences. These are spelling differences. So it turns out that this your DNA sequences are made up of a very simple alphabet. There are only four letters. At least that's the way we symbolize them. A, C, G, and T.

And here I show for example a string of A's and C's and G's and T's, and an occasional difference. Occasionally there's a spelling difference that distinguishes your DNA from that of your neighbor. 1 in 1,000. There's basically spelling difference about 1 in 1,000 letters.

Well, multiply that by 3 billion, makes for 3 million differences between any two people. So now you can look at your neighbors, even though they are also MIT alumni, and recognize your individuality at the level of your DNA sequence. Except of course, if you're an identical twin, that's a special case. You have the same DNA sequences.

Well, in the last 10 or 20 years, the field of human genetics has developed general methods for finding the spelling differences that are critical to certain diseases. I'll just give you the example here of cystic fibrosis-- a devastating disease of childhood that I'm sure that many of you have had direct or indirect contact with.

Now I'd like you to study this slide in great detail. Over here is a long string of A's and C's and G's and T's. This is part of that 3 billion letter text that each of you has.

And it turns out there's a particular three letters that are blown up over here. CTT. Well it turns out that if you're missing those three letters, CTT, you will develop cystic fibrosis. If you have those three letters, you won't get cystic fibrosis.

And so it's those three letters. The rest of the text can be perfect. But if those three letters are missing, you will develop cystic fibrosis.

Now I want to sort of give you a sense of the scale of the issue here. And I'm going to make an analogy between all of our DNA sequences and the planet. I want to give you a sense of what it means to home in on those three missing letters.

So again, our field has developed over the last two decades a powerful means of homing in on the critical spelling differences. The planet here would be analogous to the 3 billion letters of DNA that are found in each of our cells. Now we're going to move in from the 3 billion level to the 30 million letter level.

It turns out we'll now be on the appropriate chromosome. We will have found the chromosome on which the cystic fibrosis defect lies. We'll move in now to the 3 million letter level. And we can see the shoreline of Lake Michigan coming into view.

Now we'll move down to the level of about 30,000 letters. The marina is coming into view. It turns out we're now within what is called the cystic fibrosis gene. We haven't yet found the critical spelling error within it. But we're in the right functional unit.

And finally, to get down to the level of the critical spelling error, it's analogous to seeing the hairs on the back of a hand of somebody who's sitting on a bench next to that marina. Well, the point is that this is now a routine matter in human genetics. To start with the planet, to start with all of our 3 billion letters, and to end up at the end of what is increasingly a short doable thesis project for a PhD student, to end up with knowledge of the critical spelling error underlying one or another human disease.

So there are a growing number of common diseases for which certain spelling differences are known to cause, to be predisposing or causative. Breast cancer, colon cancer, heart disease, Alzheimer's. I could go on and on and on.

The hard part is finding the spelling-- I said it's become easy. But it still requires some effort to find the spelling differences underlying a disease in the first place. Once you're confident that that spelling difference is a critical predictor of a particular disease, or even when you suspect that spelling difference might be a good predictor, then it is relatively trivial, and will undoubtedly become irresistibly cheap to test lots of people for that spelling difference.

Most children with cystic fibrosis have precisely the same spelling error. So you could envision testing everyone for that or other spelling errors. And that's genetic testing. Testing individuals for critical spelling differences in their DNA.

Well, I've mentioned already you have 10 trillion cells in your body. And they all-- each of the 10 trillion cells in your body contains the complete blueprint. So if you could spare just a few cells, maybe on the way out we'll be collecting samples. If you could spare a few cells, maybe the skin cells that you left behind on your napkin at breakfast, that would be perfectly sufficient. We'll prepare DNA and get the test results back you by the end of this alumni gathering.

What would you most like to know? Would you like to know if you are among the 60 people in this room-- one in 20-- who are carriers for the cystic fibrosis spelling error, the omission of those three letters? Perhaps you would like to know whether you're one of the 200 people in this room who are three times as likely as the rest to develop Alzheimer's disease. Or perhaps you'd like to know whether you are among the 10 people in this room who have a particular spelling error that put them at a 90% risk of developing colon cancer.

Well, we could tremendously accelerate the process of finding the causes of disease if we had on file detailed satellite photos down to this level of detail, showing not just the back of a hand on a person on Lake Michigan shores, but of everything on the entire planet. If we had on file the entire repertoire of DNA in our cells, think of what we might do if we had absolutely comprehensive maps.

Well, that is exactly what the Human Genome Project is about-- to make those maps available for the study of all human diseases. Now the Human Genome Project is a federally funded effort begun in 1990 with the goal of coming to closure in just five or six years, by 2005. The goal of the Human Genome Project is to make those maps available for the study of all human diseases.

I just want to give you a 30-second summary of where we are today in the Human Genome Project. Well, it turns out here's our ultimate goal-- to know the complete string of A's and C's and G's and T's. The whole 3 billion letter text.

And there were a number of intermediate stages through which the whole field understood it had to pass. It turns out we have now passed all the intermediate checkpoints. In fact, the first complete so-called map of the human chromosome was of the Y chromosome. And it was completed here at MIT in 1992.

Three years later, also here at MIT at the Genome Center, there would be completed physical maps, there were completed physical maps of all of the human chromosomes. So it is now time to get on. Basically all these intermediate stages have now been reached. It's time to get on now with the actual determination of the string of A's and C's and G's and T's for the entirety of the human genome, the ultimate map.

And the tools to make that possible-- and in fact, the effort is beginning. Those tools are being assembled. And the effort is beginning to a very large degree here at the MIT Genome Center, as well as several centers elsewhere around the world.

I think it's very likely that we will, in fact, have a rough draft of the text. That is, maybe a 99% accurate draft of the text within 12 to 18 months. And in all likelihood, we will have a reference grade text available in three years' time. That's going to be an extraordinarily powerful tool that will dramatically accelerate the revolution in human biology.

Now you can already appreciate that there are dangers that lurk ahead as well. We're becoming good-- fast, even-- at discovering the causes of disease. But there tends to be a lag to development of rational therapies, therapies that are based on our knowledge of the causes of disease.

And in the interim, when we diagnose predisposition to disease but cannot cure, that information, if misused, could affect one's access to employment, education, insurance, and so on. And so at the same time that we recognize the extraordinary potential of this information, we also recognize that we've got to collect and safeguard this potentially toxic knowledge of our own genes.

And actually, this afternoon at 3:00 I'll be chairing a panel discussion where we're going to try to explore some of the ethical side of human genetic information. But in the last few minutes here, I'd like to turn to quite a different topic. Well, actually it has a lot to do with human variation. Maybe it's the most fundamental aspect of human variation.

Well, there's one aspect of human variation that I would claim that we have all been studying intensely since childhood. One of the oldest questions in biology. Boys and girls, men and women.

Well, it's been obvious that the members of our species come in two fundamental forms-- male and female-- for a long time. Though perhaps at the time when some of you were attending MIT, that wasn't so obvious. It's actually a lot more obvious now at MIT.

People have been wondering about how you make boys and how you make girls. What's the first question that anybody asks a new parent? Is it a boy? Is it a girl? People have been wondering about what happens before the baby is born to determine whether it develops as a boy or a girl.

Well, Aristotle was perhaps the first developmental biologist. He thought that the sex of an embro-- the sex of the baby was determined by the level of excitement of the father during sexual intercourse. Aristotle thought that the hotter the action, the greater the chance that one would have a-- I never have to complete that sentence.

Well, in the 1890s, thinking was still focused. Even as late as the 1890s, people were thinking about factors in the environment. They certainly weren't thinking about genes. The concept of genes didn't even appear until this century.

So in the 1890s, people were still thinking about environment. The most widely accepted view in the 1890s was that the mother's diet during pregnancy determined the sex of the embryo. I've always wondered how they reconcile the occurrence of boy and girl twins.

There were other seriously debated theories that focused on the state of war or peace, the phases of the moon. This is actually still widely believed. Just read the National Enquirer. Economic fortunes. There are a zillion theories.

Basically in my experience virtually everyone has their own theory as to the true cause of whether a baby is born a boy or a girl. But at least for human biologist, things clarified tremendously in the year 1959. In 1959, it became clear that whether a baby was born as a boy or a girl depended on one chunk of DNA. And that chunk of DNA is the Y chromosome.

Turns out that most of the chunks of DNA that we all carry-- these chunks are called chromosomes-- most of the chunks are absolutely identical between males and females. But there's one chunk of DNA that is found in males and not found in females. And that is the Y chromosome.

This is the whole collection of these DNA chunks, which we routinely call chromosomes. And the one with the red circle around it is found in men whose sex chromosomes are an x plus a y as opposed to in females, there are two X chromosomes. And we've come to understand that whether an embryo develops as a male or female depends very simply on if the Y is present in the embryo, that embryo will develop testes and is a male. If the embryo does not carry a Y chromosome, the embryo will develop ovaries and is a female.

Well, that's sort of what we understood in 1959. I want to fast forward now to the present. We now are coming close to understanding the complete text of the Y chromosome. And what I want to suggest is that probably by the end of this year, or at the latest by the beginning of next year, we will know all of the spelling differences that distinguish boys and girls. That will be an interesting text to decipher.

Let me tell you what we know about that text already. This was the thesis work of an MIT graduate student, Bruce Lahn, defended his thesis about a year and a half ago. And what he came up with was a gene map of the human Y chromosome. Turns out that we can recognize-- a few minutes ago, we were homing in on the cystic fibrosis gene and a spelling error within it. Well, it turns out these are a series of genes that are found along the length of the Y chromosome.

Well, talk about spelling differences. It turns out that it's not even the entirety of the Y chromosome that determines whether an embryo develops as a male or a female. It's actually just one tiny bit of the Y chromosome that makes the difference. It's actually just one gene over here called SRY. So if you as an embryo carry that one gene, you will develop testes and as a male. If you don't carry that gene, you will develop ovaries and as a female.

And to return to the concept of spelling differences, within this gene, which is of course composed of A's and C's and G's and T's, that simple alphabet that I described to you, if you carry this gene, but if you have one misspelled letter, you will develop as a female, not as a male.

So ultimately the difference between developing as a male or female can come down to one spelling change within our 3 billion letter text. That's the kind of precision with which we can now understand human development and human variation. And I would argue that's in fact it's because of that sort of precision that we now require that all of your future alumni study this subject.

So we've got this one gene that determines whether you develop as a male or a female. Well, in the last two years we've come to understand some other roles of the Y chromosome. Infertility. Infertility is an extraordinarily common problem. It's thought that about 10% of American couples who attempt to have children are unable to.

And if we went back a couple of decades, it was thought that virtually all infertility-- the finger which usually pointed at the wife. Well, it turns out about half of infertility traces to the man. And what we've found in the last few years is that the Y chromosome is a frequent site of misspellings that result in infertility. In fact, we've now come to understand that sort of going back to this gene map that I was just showing you, it turns out there are three different regions of the Y chromosome-- one here, one here, one here.

If a man is missing-- oh, and notice, remember where the sex determining gene is over here. So this part of the Y can be perfectly intact so that the embryo develops testes and as a male, a boy emerges, grows up. But as an adult, he might be unable to make sperm.

And what we found is that many men who are unable to make sperm are unable to do so because they're actually missing either this piece of the Y, or they're missing this piece of the Y, or they're missing this piece of the Y. So we've come to understand a second aspect of the Y's role in contributing to the ultimate differences between men and women.

Since we're running short on time, I think I'm going to stop at that point and just remind you that hopefully by the end of year, by the beginning of the next, we'll know the complete text-- the complete sequence of A's and C's and G's and T's of the Y chromosome. In a sense, at that point we will know all the genes, all the spellings, that distinguish the male from the female.

And with the increasingly coed complexion of the MIT student body, I think you'll agree with me that that will be an interesting text to decipher over the next years. And I'll close there. Thank you.

[APPLAUSE]

MODERATOR: While our three speakers are coming up and being seated on the stage here to handle a few questions, I want to tell you two things. One is that I was once thumbing through Dave Page's resume, and discovered that he had an honorary degree conferred on him by Swathmore. Now that is a wonderful honor. But frankly, many of our great faculty have honorary degrees. His was awarded while he was an assistant professor.

And the second story I wanted to tell you as I was listening to our first two speakers, to Martha and then to Bob Langer, there's a true and very wonderful story that ties together MIT, its effect on plasticity of the brain, and delivery of insulin and so forth. Our late great president Jerry Wiesner used to tell the following true story.

As you all know, he suffered a very debilitating stroke later in his life. And he was a diabetic. And Jerry woke up on the hospital gurney with no idea what had happened to him, but knowing he desperately needed to have some orange juice. So he decided that he would try to tell all these nurses and doctors running around that he needed orange juice. But the only word you could say was "education."

So he said, I was lying there yelling "education." And suddenly I realized what a lifetime at MIT does to you.

We have a few questions that have already been passed up from the audience. And more will be coming. We obviously can't handle all of them.

But let me begin with one that I think is very interesting. And I'd ask for very brief responses. Gets to the heart of things.

Would each of you comment on the interaction between research and teaching in your work? David, you want to start that end, and move across?

PAGE: Teaching and research. Actually, I would say that for me, the interaction has worked very much in both directions. There were branches of biology that I did not know well when I arrived at MIT. But which-- actually it was a whole area of developmental biology that I had relatively little experience in when I arrived at MIT.

I was called upon to do some teaching in developmental biology. I've since come to regard myself eventually as a developmental biologist. And so it's not just the interaction with students, but it's actually the new fields that the teaching forces me to discover that lead to my own professional redefinition.

MODERATOR: Martha?

PATON: Well, I think there are two things. One is that you yourself-- and every teacher will say this-- never really learn anything until you have to teach it. There's nothing like a group of students who know they're smart to make sure that the teacher knows his or her stuff.

And then the second thing is that as a scientist, it's the students who not only participate in the labs and do some of the experiments with their own hands, but they're the ones who notice things because they're fresh that those of us who have been at it for a long time very frequently miss. We go in with preconceived notions.

And the people will come in from different backgrounds point things out to us that we would never have learned.

MODERATOR: Bob.

LANGER: Well, I think what the last two speakers said I would only add to. I mean, I agree with both of those comments. And I'll just maybe make two additional points.

One, in some of the classes we teach today, we actually use some of the modern research as an example. So we teach courses, for example, in engineering and biotechnology and also in integrated chemical engineering where some of the material I just used today, we make central to the class. And I think students really enjoy that, because it's more on the cutting edge.

I think the second point I would make is that one of things, of course, that we all do is spend a lot of time training graduate students. And in that way, you see the research and teaching intersect, because certainly one of our objectives is to train graduate students to be the leaders of the future.

MODERATOR: Here's one addressed to Martha. It's very general. But I think just sort of some headline level comments would be very interesting to the audience. How close are we to understanding memory?

PATON: We actually do understand some forms of memory already. We can tell you how a nerve cell-- why a nerve cell responds differently to a stimulus that it's gotten a lot of before. Down to the level of what molecules change.

Now the catch is, we can explain that in terms of what happens for five minutes, 20 minutes. We can't tell you why those changes are prolonged for hours or days in some regions of the brain. On a broader level, we know what regions of the brain are not involved in memory.

There's still a tremendous amount of debate about how many regions of the brain are actually involved in memory. And so I think conservatively to put a number on it, we're not going to truly understand memory for probably another 30 years. We'll understand the molecular players. But we're not going to understand how they all work together to make the system work.

But it's coming very close. Imaging techniques where you can actually look at humans who are learning and remembering things, combined with the ability to do similar studies on animals where you can go in and manipulate, are improving our understanding monthly.

MODERATOR: Thank you. The next question I'm going to use a chair's prerogative and answer myself. Although if my colleagues disagree with me, they can.

Would more money spent at the basic research level in universities help to reduce the huge amount of money we spend annually in the USA on health care-- about $1.5 trillion. The answer is yes.

[LAUGHTER]

And seriously, there are actually some rather sophisticated economic studies that show precisely that there is a deep correlation between the advancement of basic knowledge and life science and the prolongation of life and quality of health, and ultimately cutting down on some of the costs.

Next question I guess for Dave. How soon will it be practical to get my own complete 3 billion letter gene map and compare it with others, statistical samples, correlate it with others, and so forth?

PAGE: Okay. So how soon can you know all of your own-- have a little bar code of all your own spelling differences? It's clear that technology will be in place, the spelling differences will be catalogued to make that very feasible within 10 years. I think a big question lurking beyond that will be who's going to help you interpret it?

MODERATOR: And I think you'll get into some aspect--

PAGE: That's this afternoon's discussion, if anyone would like to join.

MODERATOR: Come on if you want to know more. The next one really sort of aimed at Bob. But I think that all three of you will have some interesting input.

What are the prospects for neurotherapeutics, neuroengineering, broadly speaking, development of a new field of neurotechnology?

LANGER: Well, I'm not sure I'm the best person to address that. You may be. But I mean I think that certainly in the areas that I talked about, I guess the hope would be new treatments, for example, based on tissue engineering for Parkinson's disease. There's some work going on in that.

Really any type of cell transplantation approach. The same type of thing we talked about for nerves. I think in the drug delivery area, it's a question of drug delivery can help. But I think it's finding the right drugs, as well.

So we've already picked the example of brain cancer. And I think that that's certainly an area where more of it may go on. But I think they can really aid in any area if one's able to find better and better drugs.

MODERATOR: Thank you. Any other of you wish to comment on this at all?

PATON: Well, I'd just like to say that many people have great hopes for slow-releasing polymers when it comes to trying to remedy some of the diseases and dysfunctions of the nervous system, because there you basically have to release within millimeters a potent chemicals that you couldn't release throughout the brain. It would cause convulsions, all kinds of complications.

And we are getting very, very good at implanting things in brain regions. And so this kind of technology that allows you to restrict things is going to be tremendously potent as we learn more of where we have to put things and what we have to put in.

MODERATOR: Martha, continuing a minute. A question I'm sure everybody wants to know a little bit about. I'm kind of combining two of them.

At what age do human brains become less developmentally plastic? Is it different for different kinds of stimuli-- visual, audio, and so forth.

And then there was a second question that really asked what general conclusions we should be drawing for education and raising of young children from your work.

PATON: The last one is-- I'll save it. So yeah. The difficulty is that different neural systems mature and lose plasticity at different rates. And some of these interact. So you know the nervous system isn't just a nice linear array. There are feedback systems.

And so one has to first understand what's the rate-limiting pathway for a change in plasticity. Different for humans, different pathways mature at vastly different rates. Newborn, the somtasensory pathway is supremely tuned. And in fact, contact is exceptionally important for young children. The auditory system is still tuning. And the visual system, even though the peripheral apparatus is working fine, the central nervous system still has a tremendous amount of tuning.

So these problems with squint and strabismus that I was talking about, you have to correct them within the first three years of life to not have any impairment. And it varies from individual to individual. There are individual differences. And yet other auditory pathways retain considerable plasticity at some levels later.

I gave you the example of the American Sign Language. That is pretty solid data, because they can do it on normal developing children. And you can show that there really is a discrete cut off around four years of age for these massive brain changes that are associated with using motion and vision in a language context.

And then to the last question. I think there are tremendous implications of this developmental plasticity for how we deal with our children as a society. It's I think one of the most important things that biologists working in this area can do now. And that is to convince people that the need to have good early education, stimulating environments for young children-- not at five, when we classically started. But actually early. As young as they can spend time away from their mothers.

It's just a historical accident that we start education-- formal education at five. And there are lots of longitudinal studies that say it should be done earlier. But unfortunately, there are political reasons why that money doesn't want to be spent in that way. It would probably save us a tremendous amount of money in the later years of life.

There is a lot of work. Children are highly plastic. They're capable of absorbing tremendous amounts of information from toddler hood onward. They learn a lot from other children and from dealing with a range of different kinds of different adults in different situations. And we should be paying a lot more attention to early education and to childcare.

MODERATOR: Thank you.

[APPLAUSE]

MODERATOR: Switching gears and tossing a question to Professor Page that even I know enough about to know really is loaded. But it's a great story. What role do private companies like PerkinElmer-- can't quite read all these. But it says human genome. I think we know what we're talking about.

What role do private companies have in the sequencing of the human genome. And are they competitive or collaborative?

PAGE: All right. Okay. That's a very hot question. Actually, just yesterday I spent the day at NIH reviewing grants. And it's always the topic in the background.

It turns out the Human Genome Project has, as I said it, it started as a federally funded effort officially in 1990. It has succeeded wildly. It's been ahead of schedule, under budget, and of course, there were really two motivations-- sort of selling points for the Human Genome Project early on.

First, it had great promise of improving human health. And I think that will be realized. And secondly, it was realized that biotechnology was an area in which the United States is preeminent. And that pushing the Human Genome Project would help maintain that preeminence of US industry.

Well, that has come to pass with such lightning speed that now genome science is a booming area in the private sector. And in fact, the power of the information, the power of knowing the text, the 3 billion letter text, is so widely appreciated that private investors would like to have a look at that information six months before it becomes publicly available.

This is perfectly serious. And so there is a race on now. I mean, to be honest, there is a race on between the publicly funded sequencing the human genome and privately funded sequencing of the human genome, recognizing the extraordinary power and the extraordinary intellectual property that resides within that sequence.

There is hope that this effort will be collaborative. I think it will be like all of science has always been. That is, a strange mixture of collaboration and competition that will drive the field forward. But it's going to be an extraordinary-- the next 12 to 24 months will be extraordinary.

MODERATOR: Thank you. Bob Langer. Who is working on growing whole organs using the polymer techniques, or what's the stage? When is it going to be ready to go?

LANGER: Well there are a variety groups working on areas like this. There's work going on at Children's Hospital with some of our collaborators led by Tony Atala, a urologist trying to make a bladder. There is research going on where people are trying to make a liver. Linda Griffith, one of my former postdocs, now a professor at MIT, has been working on that.

I'd also add that there are companies-- like there's a local company called Circe Biomedical that's working on a liver. It's what's called a Bridge to Transplant. I that showed that picture of the small child who was dying while awaiting a liver transplant. And right now they're in the final stages of clinical trials using systems that could keep a patient like that alive for several weeks while awaiting for a transplant.

MODERATOR: Martha, is there an implication about the incidence of attention deficit disorder to be inferred from the research on developmental plasticity of the brain? For example, does excessive stimulation at an early age result in a later inability to discriminate among stimulatory inputs?

PATON: Unfortunately, we don't understand all the pathways involved in attention deficit disorder. It's a disorder that we infer is associated with the transmitter dopamine, because you can treat it by pharmacologically altering the dopaminergic system.

But like many things in the brain, we don't understand really why that works or why it occurs in the time scale it does. And there are neurological diseases and disorders that have to do with not trimming these exuberant projections I was talking about. For example, many people believe that some aspects of autism are associated with it. So if you don't trim them appropriately, that's not good.

But it's also becoming quite clear that a lot of this plasticity, this example of the owls that can actually achieve this adjustment later in life if they are asked to do it early on, that is probably the result of keeping some exuberant projections, but keeping them suppressed until you need to use them. So it's a balance of both this excitatory drive and selective use of inhibitory systems.

And I didn't talk about that. An attentional deficit disorder has all the attributes of a failure to use inhibition appropriately. We have no idea of how many different areas of the brain are involved when that goes wrong. And it may well be that attentional deficit disorder defined as a syndrome, it may not be that it's the same in all people who have been diagnosed that way.

MODERATOR: Thank you. I'm going to give two questions to David. The first, which is headline stuff at the moment, Dolly the clone has aged. I can't quite read it. But as we all know, she appears to be aging much more rapidly possibly because she was cloned with mature genes. What do you think is the impact of this knowledge?

And secondly, while we're talking about these things, how is gene therapy executed? Obviously the genes in every cell can't be changed, so the card says. It's a little bit about technology, a little bit about the recent Dolly aging.

PAGE: Okay. So first let me deal with Dolly. Actually our now missing panel member, Bob Weinberg, had actually commented extensively on this in the press in the last week. And Bob correctly pointed out that there has been extraordinary press attention to what looks to be the aging of the DNA chunks, the aging of the chromosomes in Dolly.

The present scientific evidence is a mere shred of evidence. And it's far too early to draw any conclusions about whether Dolly's chromosomes really show signs of aging. But I think that the attention drawn to this shred of evidence tells us more about the extraordinary public interest in the question than it actually does about the scientific data that's available at the moment.

With regard to how gene therapy is done. I think it's-- first, put it in broad perspective. Gene therapy is not yet a reality. No one has been cured of a disease by gene therapy as yet.

But the question was a very astute one. That is, if you've got these 10 trillion cells, and they've all carrying the blueprint, how do you travel from each of these cells? How you travel among all 10 trillion correcting the spelling error?

Well, it turns out that many of the spelling errors have consequence only in a particular part of the body. So in the case of cystic fibrosis, for instance, many of the most devastating complications of cystic fibrosis involve the lining of the airways to the lung. They get clogged.

And so in the case of cystic fibrosis, for example, there is no gene therapy at present. But it is being attempted. And the attempt is being made to deliver the good gene, the properly spelled gene, to deliver it directly to the airways of the lung.

And in that case, you've all had foreign invasions of the airways to your lung. You've had colds. And in fact, people who are developing gene therapies for cystic fibrosis are actually trying to piggyback the good gene onto the properly spelled gene on the back of the bugs that cause colds. So that's an attempt to harness the vehicle that's appropriate to the target. And there are many other similar stories.

MODERATOR: I'm going to give a couple of questions for sort of a quick reaction to Bob Langer. First question says, might a member of the class of 1967 live to see his arthritic knee cartilages replaceable? And the second is also a serious personal question wanting to know about how one would go about getting information about new uses of artificial skin. In this case, the person has had skin removed from one part of his or her body to repair another, and wants to know sort of where that stands.

LANGER: Well I guess the answer to the first question is, I hope so. But you know, it's hard to really predict when medicines are going to be available, or when new therapies are going to be available. And the last thing that people like myself who work in these areas want to do is create false hope.

I think that there's very active work going on. There are clinical trials ongoing in different situations. But I think the very key thing is just because they're doing it doesn't necessarily mean it will work. And my hope is that within 10 years things either will be available, or we'll know some good answers to it. But you really can't be sure about the timing.

The second issue about the skin, there are several companies making new skin. And probably the best thing to do is to call one of them. There's a company that I've mentioned doing skin work that's based on some of our work. It's at Advanced Tissue Sciences in La Jolla, California. There's another local company called Organogenesis in Canton, Massachusetts. And there's a company called Genzyme tissue repair in Cambridge.

And each of them have different types of approaches for making skin. Probably the best thing to do is to contact people in those companies, I would think.

MODERATOR: Thank you. Now Martha, because you're about to become a new member of the MIT faculty, our alumni and alumnae are being very gentle and not asking hard questions today. Are you ready?

PATON: I'm ready.

MODERATOR: Please define consciousness and tell us what we know about it.

PATON: This is very unfair. I don't think neuroscientists have any better definition of consciousness than poets, actually.

[APPLAUSE]

We still debate where down the phylogenetic line consciousness starts. It clearly involves the cerebral cortex. But the cerebral cortex requires activating things from the brain stem to achieve it. So it is a state in which the cortex has a certain pattern of electrical activity.

However, that's not the defining factor, because in many times in a sleeping brain you can have similar patterns of activity. And are you conscious when you're dreaming? So I have been at meetings of neurobiologists, actually planning meetings for the NIH, where we spent in this panel of experts an hour and a half trying to come up with guidelines for studying and defining consciousness. There isn't a good answer yet.

MODERATOR: Are there any poets who would like to comment on these questions?

David, another sort of question like that that we all think and try to understand. Since twins are born with the same genetic code, what happens at the cellular and molecular level that makes them different? Even if the environmental-- can't quite read that. But they're asking really about the environmental differences that cause twins to grow up being very different people. And do we know anything about what occurs down at the cellular and molecular level you worry about that feeds into that understanding?

PAGE: Well, I think I'm going to twist my answer around a bit. I think that's an invitation to speak to the degree to which we are a simple readout of our genes. I mean, as I mentioned, identical twins have no spelling differences between them. They have the same instruction book.

I think it's very popular these days. Our media are full of stories about the gay gene, the happiness gene, the gene for anxiety. There's an idea is a field that we are just our genes and nothing more.

I think nothing could be further from the truth. Actually it will take decades, it will take hundreds of years, it will take forever to figure out where the actions of our genes stop and where the actions of our parents and our environments and such begin and where they stop. And so it's a perfectly framed question, actually.

Identical twins are in fact the context that human geneticists most like to examine, because the huge differences that we know can arise between identical twins we suspect are a reflection of the complexity of the environments. And even though we say identical twins, they may grow up in the same household with the same parents, their experiences are never exactly the same.

But it really brings me around to this very fundamental question. What are the limits of genetic determinism? And I think we don't know them. But I think they're very frequently overstated.

MODERATOR: Yes, Martha. Please.

PATON: I just want to add something. There are beginning to be some very intriguing studies now of genetic monozygotic twins, identical twins who have different behaviors. And there was one very interesting study of twins afflicted with Tourette's syndrome, which has a genetic component. It's a motor disorder. And it has various degrees of severity. Some people function quite normally. And some are very severely afflicted. They have personality disorders and tics and things.

And there were a set of imaging studies, this functional magnetic resonance imaging studies done. No actually it was a different type of imaging. But they were looking at the distribution of a certain population of receptors in regions of the brain that control these tics. Regions called the caudate nucleus part of the basal ganglia.

And they had a panel of neurologists evaluate the severity of the disease in these twins. And then completely independently they imaged them. And they correlated quite precisely distribution of some receptors in the caudate nucleus with the severity of disease.

They were raised in the same home. These are monozygotic twins raised in the same home. And there were organic differences in their brain that showed up in behavior that correlated very closely. So clearly it wasn't just the genes in that case.

MODERATOR: David, you mentioned the press on the gay gene as you were answering that question. There is a specific question here asking if there are correlations between the gene alphabet and homosexuality.

PAGE: Ah. I knew that question had to come. That's why I was hoping that perhaps I dealt with it in my previous answer. But I should've known better.

This is actually the question that I'm most frequently asked by both my parents and my in-laws when the topic of my work comes up. To what degree is sexual preference encoded in our DNA spellings? The answers are far from in. The first suggested evidence appeared in 1995, I believe.

There is reasonably strong evidence, though controversial, there is reasonably strong evidence that homosexuality in men is at least partially predisposed to by spelling differences. But that evidence remains very controversial. It's not clear how significant a factor it is. And even if it is a factor, it leaves open the question of the role of environment.

So I guess where I'll end up is if I had to bet today, I would say yes, there probably are genetic predisposers to sexual preference. But the story is very, very far from complete.

MODERATOR: Thank you. I'm going to end by giving everybody an opportunity if they wish, each of you an opportunity to comment on this rather general and interesting question. Do you think that the enhancement of life and treatment of disease in the future will primarily come from natural substances, artificially generated drugs, or DNA technologies? Anybody care to comment on that?

By the way, we have some faculty here who have very strong views on this. But maybe it's not among these three.

PAGE: Begin wherever you'd like.

MODERATOR: Bob, start with you. You're the engineer.

LANGER: I'd say the last two. And I think that they may work together, I think. I think DNA work gives you targets. And certainly genetic engineering has been a powerful tool and method for generating the drugs. And again, different types of approaches that people are using for drug design, I think, are getting more advanced. And I think we'll see more and more drugs made that way, as well.

PATON: I'm no expert in this. But from what I read, a lot of companies have teams that go out, particularly into the rain forest, because they think these things are going to disappear. And they dig up soil bacteria or organisms that live in the high canopy. Or they go down to the depths of the ocean. They pull things out of hot springs to find natural substances that they can then use on tissue, and figure out what about those molecules is making them effective.

So I think the answer is all of the above. I mean, one of the molecules that I know many of you will appreciate that has made DNA technology possible is an enzyme that can function, that isn't destroyed at very high temperatures. And it comes from a bacteria that is a hot spring-- he should be answering this-- that is adapted to live in tremendously hot places. So it needs these enzymes that can function at those temperatures.

PAGE: I'm going to answer the question in sort of two parts. I think it's worthwhile remembering what the major causes of disease were in this country earlier in this century. And those were infectious diseases. That remains the case for the majority of people in the world.

So if we actually look beyond our own borders, I think the question as to what will improve human health and well-being, it will actually be a combination of economic development, improvement of sanitation, and focused battles against the parasites and infectious agents that still plague the majority of our species. Within our own more highly developed societies, I suspect that the most important improvements to health and well-being will actually come from ongoing changes in human behavior, smoking, and then from what I was vaguely alluding to in my presentation. That is, the ability to customize preventative health care based on knowledge of our predisposition.

MODERATOR: Thank you. As we end here, I just wanted to point out that inevitably I've had to bypass a large number of extraordinarily good questions, mostly for three reasons. One, several of the points were actually touched on as our panelists answered other questions. Two, there were a large number which clearly were aimed at Bob Weinberg's talk. And unfortunately, as we knew in advance, Bob would not be able to be with us for the Q&A. And three, there are some remarkably interesting questions that are going to fit precisely into the afternoon session that David mentioned, dealing with some of the ethical and moral issues that we now are going to be confronting, running all the way from some of the genetic technology-based questions to things like bias of medical research toward males in the United States and so forth.

So there's going to be a lot of food for thought. And I hope that all of you, one way or another, will get a chance to address your questions to these or other experts as the day progresses. We are now going to break up in just a moment and move as quickly and efficiently as possible to the Johnson Athletic Center where we will reconvene for technology day luncheon. But please join me in welcoming Martha to the MIT faculty and in thanking all three of our colleagues. Thank you.

[APPLAUSE]