Teacher resources and professional development across the curriculum

Teacher professional development and classroom resources across the curriculum

Monthly Update sign up
Mailing List signup
Search
Follow The Annenberg Learner on LinkedIn Follow The Annenberg Learner on Facebook Follow Annenberg Learner on Twitter
Rediscovering Biology Logo
Home
Online TextbookCase StudiesExpertsArchiveGlossarySearch
Unit 2: Proteins and Proteomics
Back to Unit Page
Unit Content
Unit Textbook Chapter
Animations & Images
Expert Interview Transcripts
Unit Glossary
Related Resources
Unit Activities Back to Transcripts
Leroy Hood, PhD

Leroy Hood, Ph.D.
Hood played a key factor in the Human Genome Project and was credited with developing the automated sequencer. He is recognized as one of the world's leading scientists in molecular biotechnology and genomics. Hood created the cross-disciplinary Department of Molecular Biotechnology at the University of Washington, bringing together chemists, engineers, computer scientists, applied physicists, and biologists. He has recently created the Institute for Systems Biology in Seattle, Washington, and serves as its president. Recently, Hood's lifelong contributions to biotechnology have earned him the prestigious Lemelson-MIT Prize for Invention and Innovation.

You played a key role in creating sequencing technology so we can examine the smallest parts of an organism. Now it looks like you are working in a different direction. Can you talk about that?

Well, the DNA sequencing technology that we developed in 1986 made it possible to do the genome project. And that will be finished, theoretically, in April of 2003.

What the Human Genome Project has done is given us a genetic parts of all the components that make up a human and make up many other organisms whose genomes have been sequenced as well. And thus, the genome has enabled a completely new approach to biology that we call "systems biology." And the idea, put quite simply, is that for the first time we can take a biological system like the immune system and we can look at virtually all of its components and see how they behave with respect to one another. And it is in this way, by analyzing globally all of the components, that we'll truly come to understand systems.

Biology for the last 30 or 40 years has primarily focused on looking at one gene or one protein at a time. And although that's gained insight into biology, it hasn't really gained any insight into how very complicated systems work.

To give you a sense, one example: we make vaccines today exactly the same way we made vaccines a hundred years ago, even though we know more about the genes and the proteins of the immune system than virtually any other human system. And the point is, until you use these systems approaches, you'll never really understand how systems operate and how their systems properties emerge.

Systems biology is an approach which must integrate many different disciplines. It takes some chemistry and computer science and engineering and biology. It's an approach that looks at the entire system. So if we're interested in the immune system, we have to be able to study all of the components of the immune system.

We have to study [the components] at many different levels of information. We have to study its DNA, its RNA, and we have to study its proteins and how they interact. It's only by studying each of these levels of information-integrating them all together-that we'll be able to get a coherent picture of the immune system and how it actually functions.

So that's what systems biology is about. It's using new technology and all the insight that's been gained from the Human Genome Project, and most important, it's developing powerful new computational tools to handle all this information and integrate it together and ultimately create mathematical models which can actually predict the behavior of the system given any particular kind of input.

Can you talk about the importance of protein networks and interactions in the understanding of organisms?

Well, just as the DNA in the genes are essentially the information for making organisms, the proteins are the information that manifests itself in the organism. When you look at another individual, virtually everything you see is proteins. And all the biochemistry and all the thinking in the brain, and many other kinds of things-most of these processes are mediated by proteins.

If we're ever to understand how complicated systems work, we need to be able to understand the networks of proteins that are involved in the immune system, or in the nervous system, or in the heart. [We need to understand] how they work in a coordinated fashion to enable those systems to carry out their functions, be it responding to foreign pathogenic organisms, be it learning how to think and remember, or be it pumping blood around the body so that you can carry food and oxygen to all the cells.

Can you explain what a "predictive model" is?

A "predictive model" is an attempt to formulate a description of how a system behaves. A predictive model can be descriptive-that is, it can be written out as words; it can be a graphical model displayed in two or three dimensions on a computer; or ideally, it can be a mathematical model that precisely describes the behavior of the system.

But the idea of a predictive model has essentially two different components. One is that given any kind of input to the system, you can predict exactly how the system will behave. And number two, if we understand enough about the system, we can actually [redesign] it to create completely new kinds of properties. So predictive models, mathematical models, are critical for this level of understanding, and this level of understanding is really the key for medicine in the future.

One benefit of having a predictive model is it is going to push us towards a completely new type of medicine which we call predictive, preventive, and personalized medicine.

In the next ten years or so, we'll know enough about defects in your genes and how they cause disease that we'll be able to take a small snippet of blood and analyze the genes in that blood, and then be able to write out a predictive health history for you.

Whether you have a 50% possibility of heart disease, and a 20% possibility of diabetes, and so forth. So for each individual, we'll be able to write a probabilistic health history. And at the same time, we'll have developed little hand-held devices using a technology called nano-technology that will have the ability to make just a small pinprick in your thumb, take a small amount of blood, and analyze 10,000 components that are in the blood. From those components, we'll be able to determine your health status. That is, are you normal and healthy, or have you started to get a particular kind of cancer or a particular kind of heart disease.

The idea would be to take the information from those 10,000 measurements and put it in a computer server that would do an analysis and send back a printout to you that said, "you're ok-do this test again in six months" or "go see your cardiologist or your oncologist." The idea for predictive medicine is not only can we make predictions about the potential of diseases that you can acquire in the future, but we can monitor the potential acquisition of these diseases and determine when they start and how far along they are.

The second part of this kind of medicine is preventive medicine. It's here that we'll be able to use these approaches we call systems biology-the defective genes in the context of their systems. And by understanding these systems, we'll be able to circumvent the limitations of these genes by making new preventive drugs. The drugs might be [the] use of stem cells, they might be modified genes, they might be modified proteins, but they might [also] be classical small-molecule drugs.

The idea is, if we say that you have a 70% chance of getting breast cancer by the time you're sixty years old, we'd like to say, "if you start taking these pills when you're forty years old, you'll probably never get that breast cancer." So the idea of preventive medicine is [that] it suggests that we'll be able to not only extend the lifespan of the average lifetime of the normal individual by ten to thirty years, but those ten to thirty years will probably be productive.

This raises two interesting social issues. One is, we don't treat older people very well, and if they're creative and productive, often society takes advantage of them. And the second is, we're going to have to totally revolutionize the education of physicians. If you think about it, if the predictive and preventive medicine comes to the fore in the next 10 to 20 years, the physicians we're training today will be practicing the kind of medicine for which they're utterly inadequately educated to [deal with].

So the idea of changing medical education is certainly going to be a central scene. And then the final point of this new medicine is that it'll be highly personalized because the genome has shown us that each of us differ by six million nucleotides on average in our DNA from one another. That means we'll be susceptible to different combinations of diseases, each of us.

What that means is physicians in a uniquely new way will have to treat individual patients as individuals-they'll have to be worried for each individual about the particular combination of diseases that they are predisposed to and muster preventive treatment for each of those. So the idea that each patient has to be treated in such an individualized manner will be a big revolution in practicing medicine.

Can you talk about the Institute's work in tuberculosis and why was this disease chosen for research?

Well, I don't work on tuberculosis myself. Another researcher here at the Institute for Systems Biology does work on tuberculosis and he works on it for two interesting reasons. First, he is an immunologist interested in one branch of the immune system called "innate immunity"-the earliest branch of immunity that actually evolved. And number two, he came from South Africa, so he's always had an interest in third-world diseases and emergence of such diseases.

In some ways tuberculosis is classified as a third-world disease. What we can do at the Institute for Systems Biology is begin studying infectious diseases such as tuberculosis using these new systems approaches which give us very, very powerful new insights not only in the early diagnosis for disease, but in thinking how to design much more rational and much more effective treatments for disease. And finally, for most infectious disease, it's pretty clear that the most effective way to treat them is through vaccine, and not through drugs. Alan, being an expert in the immune system, is investigating new approaches to vaccine that can be used for many third-world diseases-HIV, AIDS virus, tuberculosis, malaria, etc.

What are the steps in creating a predictive model?

The steps in creating a predictive model are first to take all of the biology that you know about the system that you'd like to make predictions on, and formulate a very preliminary model. The second step is to do the systems biology analysis of that system, where we look at all its genes and all its proteins and all their interactions. And we integrate this kind of information together and come to a somewhat more refined picture of what the model should be.

But as we continually refine that picture in an integrated fashion, we'll go back and do more experiments and more analyses. So we create new hypotheses, we go back and test them, we reformulate the model. So in this process, gradually the model comes to describe the reality of the system with more and more specificity.

So the basic idea, then, in systems biology is to be able to integrate new technologies, to be able to use new computational tools, and then to interface them really effectively, with the biology of the particular system you're in.

Can you provide an example of something that you have studied at ISB and run us through that model?

Something we've studied at the Institute for Systems Biology is how one creates a developmental program for generating a sea urchin. A little sea urchin is an invertebrate, and it goes from one cell to its earliest larval form with 1,800 cells in about 72 hours or so.

Across that 72-hour period, the larva makes its skin and the gut and the nervous system and the skeletal system. And what we've done is identified the nature of the regulatory gene network that specifies how the gut of the sea urchin develops. We have a network that includes 55 genes that describes precisely how, across that 72-hour period, the development of the gut of the sea urchin occurs.

In fact, we understand sea urchin development so well, that by rationally manipulating just one gene, we can create completely new types of systems. For example, we created sea urchins that didn't have one stomach, but actually had two stomachs. And you could predict that accurately from the model. So this gives you an idea of how-once we understand the regulatory development of systems-we can begin to manipulate their information for the benefit of humankind.

The reason that we study sea urchins is because they're an ideal model organism for studying something that is very difficult to study in humans. First of all, they are a relatively small creature. Second, we can get literally billions of eggs in a single summer to create many different types of sea urchins and we can synchronously fertilize millions of these eggs and stop them at any stage of development to investigate the gene programs and how they're actually running. And, third, they're a transparent organism, so you can actually put genes that color certain aspects of development into the sea urchins and you can see how those colors change with time, and that gives you additional insight into the nature of the gene regulatory network.

So the sea urchin is a wonderful tool for studying some of the most basic mechanisms that operate in all living creatures, including human organisms. One of the fundamental insights we've gained from the Human Genome Project is the fact, by looking at genes-almost now of probably 50 different organisms-is that we did all descend from a common ancestral origin and in fact, for basic mechanisms, we all use very much the same kind of strategy because we did come from that early initial common ancestor. So this lets us use many different model systems to rationalize the study of human biology because of the evolutionary relatedness.

What are the still-unanswered questions in systems biology?

Systems biology is the very, very beginning to almost every question that's unanswered. There are unanswered questions about how we will develop the new technology to be able to do these necessary global analyses of DNA and RNA and proteins and interactions. Parenthetically, one approach that we think will really revolutionize biology is to use the tools of nanotechnology that allow us to study one molecule and one cell at a time.

With the tools of nanotechnology we can make many, many measurements in a cost-effective fashion and do it on little instruments that are not much larger than the dot of a pencil. So there are many technical problems that we have to surmount. When we invented the DNA sequencer back in 1986, it could do about 250 nucleotides a day if we were lucky. And if we look at what machines today can do, they're about 3,500 times as fast in their throughput of DNA sequencing as our original machine was.

We're working on a new approach to DNA sequencing using just single DNA molecules that in 10 to 15 years may again be 3 or 4,000 times as fast as what we can do today. And if we succeed with this endeavor, we can envision the time when we can do a whole human genome in less than half an hour and maybe for a cost of under a thousand dollars. This is actually going to be the primary driver of this predictive medicine because in the future every human will have their entire genome done. And then we'll analyze the hundred thousands of genes that predispose for disease and then we can make these probabilistic predictions.

So for systems biology, one enormous challenge is, what are the technologies that we can point towards the future? And one big technology-that may be at the heart of the technologies we're working on now-is something called proteomics. That is the ability to analyze proteins in a global sense. The second area that needs to be worked is the area of computational tools--how do we effectively store and integrate many different kinds of information.

How can we actually analyze it statistically? How can we put together these models that we were talking about initially-graphically, and ultimately, mathematically to make all of these predictions? Each of these phases of systems biology is going to require enormous invention and genius on the part of our mathematicians and computer scientists and physicists at the Institute. For systems biology we have people with all of those disciplines working on these types of problems. The final issue for systems biology is how you take these technologies and computational tools and effectively interface them with biology.

That's the real question of how you bring the cross-disciplinary scientists all together. How you give them a common language to speak, how you focus them on coherent problems where you can point these very powerful approaches. The ability to integrate people, the ability to integrate technology and computation and biology and ultimately of course, will be integrated medicine into this operation, as well. I think those are the nature of challenges of systems biology. And again, in some very simple model systems, we've had dramatic demonstrations of the power of systems biology. But, it's just the very beginning.

It's about where the Human Genome Project was in 1986 or '87-we're just beginning to do experiments and think about what the needs are and how we put all these things together.

In what other ways will systems biology change the world?

Systems biology in the 21st century will be the way all biology is practiced. It's going to change how we do agriculture, it's going to change how we do animal husbandry, it's going to change how we do medicine. And, interestingly enough, it may change things like the evolution of human beings. Systems biology gives us the insight and the knowledge so we can think about doing germ-line engineering in the future-change in the positive sense.

Traits like intelligence, or physical attractiveness, or physical ability are questions society is going to have to face. Should humans take the engineering of evolution into their own hands and start creating, not leaving to the random chance of natural selection and mutation but actually creating new types of individuals that have these different kinds of traits. My prediction is that this will be an issue of enormous debate in society.

What do you see as the future of systems biology?

I think the future is going to be learning how to do systems biology-I think the future is going to be applying systems biology to lots of really challenging problems in biology of medicine. I see systems biology as spawning a whole series of new companies that can do specific kinds of things.

I think probably there will be a whole restructuring of the pharmaceutical industry because the big drug companies are going to have trouble doing systems biology, so companies will emerge that will specialize in the ability to integrate technology and computational biology and medicine and really effectively use that in terms of searching for targets for drugs and things like that.

What do you dream about seeing in systems biology?

What I dream about seeing is a discipline that is as productive and creative as I imagine it will be in the future. Again, it's like a new baby, and what you'd like to see is that baby grow up and be as productive and full of potential as you imagined.

Can you talk about the difference between protein evolution and evolution of an organism?

That's a very complicated kind of question. It turns out if you look at the genome, there are really two major types of information for the presence of the genome. So one type are the gene that encodes these proteins that you were asking me about. Proteins are basically molecular machines that give the body shape and form and catelize the chemistry of life and do essentially all the functions of life. And most people have thought about evolution strictly in terms of proteins and how proteins have changed. And indeed, how proteins change do give us good sign posts as to how rapid evolution is occurring in the nature of some of the changes.

But it's the second type of genomic information that really holds the key to evolution in the future. These are gene regulatory networks that actually specify the behavior of the genes themselves. And what we do know is that gene regulatory networks can change far more rapidly than the protein themselves can change. Let me just give you an example of this.

If you compared the brains of a chimpanzee and the brains of a human, they're really different from one another. The human's is bigger, it's got many more convolutions, the density of neurons is greater. So what that means is the gene regulatory network for developing the brain has changed in just the six million years of divergence between those two species quite markedly. But if you look at the proteins themselves, they have changed less than one residue in two or three hundred. So they've changed very, very little. So my own view is proteins reflect evolution. But the driving force for evolution are changes in these gene regulatory network that basically control the behavioral genes.

Is it possible to define the proteome of an organism?

The proteome is all the proteins that are expressed in a particular cell type. So a kidney cell has a different proteome from a brain cell: to carry out their function, they express different proteins. But if you want to know what the potential proteome of the organism is, of course, you just have to go to its genome, and it can't express more proteins than that gene. That is, each gene expresses a different protein.

So in theory, one can define the theoretical proteome for an organism like a human. But in practice, what you're really interested in is what does the proteome in the kidney cell or what is the proteome in a brain cell really look like? Because those are the elements, those are the players [with which those] particular cells carry out their biological functions.

Can we ever know all the interactions of the proteins in an organism?

So obviously if you have 30,000 proteins, then in principle every protein could interact with every other protein-you'd have an enormous number of potential interactions. In fact, we know that just some proteins interact with other proteins, so the nature of the interactions are very, very constrained. And one of the key tools of nanotechnology that we believe will be developed in the future is better ways of doing protein interaction measurement with very small amounts of material in a very, very large scale. So I think we will have very powerful tools for determining virtually any of the interactions that happen in humans, but that might be ten years away from now.

How has proteomics changed biology?

I think proteomics per se hasn't changed biology. What it's done is emphasize the importance of protein for the functioning of organisms. [Proteomics] is much more complicated than genomics, because proteins have twenty different sub-units rather than four subunits nucleic acid, so they're a much more complicated identity. And proteins can be modified in the body by four hundred or more chemical alterations that change their behavior, and proteins that interact with one another and proteins that compartmentalize in special parts of the cell. So all of these different dimensions of proteins are things that we have to understand, so what we've come to realize is the study of [proteomics] is going to be an enormous, enormous technical challenge. But as we define it, we define an incredibly important set of components that are going to give us deep insight into how the immune system works, the nervous systems works, and how the cardiovascular system works.


Top Back to Transcripts

© Annenberg Foundation 2014. All rights reserved. Legal Policy