Michael Gottesman: Purely Academic

Graphic

Michael Gottesman has always approached scientific research with a long-term view. He was greatly influenced by the sense of optimism that science fostered in many postwar baby-boomers, and he has carried this optimism with him throughout his career, for the last twenty-eight years at the National Institutes of Health. Intent upon making a contribution to human biology, he joined the National Cancer Institute to address problems of basic cell division, and for his seminal work in cancer research, including fundamental advances in elucidating mechanisms of multidrug resistance, he was elected this past year into the Institute of Medicine. In addition to his role as Chief of the Laboratory of Cell Biology at the NCI, Gottesman has also directed the intramural program at NIH for the past decade. His initiatives as Deputy Director for Intramural Research have touched all levels of training at the NIH, providing high school teachers and their students with laboratory experiences on the NIH campus, expanding the opportunities for undergraduate and graduate students in intramural research projects, and more recently, reinvigorating the fundamental role that the NIH plays in training postdoctoral researchers. The optimism that gripped him as a child in the 1950s is still apparent as Gottesman discusses his research and his aspirations for the far-reaching impact that the NIH continues to have on society.


Graphic

MI: It’s been quoted that you were inspired to enter science in your youth by the launching of Sputnik.

MG: Yes, I was a child of the Sputnik era. I was eleven years old in 1957, growing up in Queens, New York, when the Russians launched a satellite and took everyone in the U.S. by surprise. I was already interested in space, as most young fellows are, when the launch of Sputnik generated a level of excitement about science that we’d never had before. I always had a chemistry set, and by that age I was also experimenting a little bit with hobby rockets. I was always headed for science.

MI: And when did you know that you were headed for a career in biomedicine?

MG: I remember that when I was interviewed for college, they asked what kind of career I was interested in, and I said that I would like to emulate John Enders, the scientist who developed the tissue culture system used for growing the polio virus that later led to the polio vaccine. Not only was I a child of the Sputnik era, I was also in that cohort that got the first polio vaccine. Millions of children in U.S. received the polio vaccine in the early 50s, and I remember being part of that. At the time it didn’t seem so exciting—it was mostly sort of frightening—but in retrospect I think that conquering polio was a major influence on people of my generation. I can remember my brother being sent home from summer camp because of the polio scare. Polio was a terrible disease in this country.

MI: When you entered college (Harvard), were you already planning to get your MD?

MG: Well, I was generally interested in medicine for its own right. I wanted to learn as much as I could about human biological systems, but I don’t think I ever thought I would practice medicine. It used to be that a standard pathway to research was through an MD rather than a PhD. And many of the people who were my heroes in medical research were MDs. When I graduated from college, there were a couple of MD-PhD programs, but I decided to stay at Harvard for my medical studies, and part of the reason was that my wife hadn’t graduated yet. She was at Radcliffe, and I was at Harvard; we had met in a physics class and gravitated to one another. She was always interested in bacterial genetics, and I think she was a great influence on me.

MI: Who else was influential to you while you were at Harvard?

MG: Another person who was important to me as an undergraduate was my tutor, Bill Beck. (William Beck died just last year.) Harvard has a tutorial system, so that everybody who is majoring in biochemistry, in addition to all the courses and professors, has somebody who serves as a personal academic guide. And a very influential part of my training was that tutorial at Harvard. We spent a couple of years reviewing Arthur Kornberg’s first sixteen papers on DNA polymerase, which were published in the Journal of Biological Chemistry, in fine detail. We were always waiting for the next paper to come out—they were labeled with Roman numerals I through XVI, so it was, you can imagine, like awaiting installments of Charles Dickens. We went through every paper, looking at techniques and poring over data.

When I was in the Medical School, Bert Vallee was one of my professors, and I worked in his laboratory during my first two summers and, later, for a six-month elective term in his laboratory. I worked actually more directly with a graduate student in that laboratory who was very influential in my career, Bob Simpson. Bob was an MD and was getting his PhD; he was a wonderful mentor and helped to guide me in terms of career. Bob eventually came to NIH and then about fifteen years ago or so moved to Penn State as Chair of the Department of Biochemistry. I’ve been fortunate to have a lot of people who were very well informed and very capable but also very interested in supporting my own career interests.

MI: Is it just a matter of luck that you got good advice?

MG: No, I sought out those people. I mean, mentorship is a two-way street: you don’t just fall into it; you have to seek it out. An important bit of advice for students.

MI: Knowing that you wanted to go into research, did you find the clinical training that you must have undergone to be a frustrating diversion?

MG: No, I liked it. I did an internship at the Peter Bent Brigham Hospital, which meant really hard clinical work. The average working day was thirty-six hours long, and then you’d go home and sleep for twelve hours, and then you’d go back to work—and that went on for a year. The thought of it now is just unbelievable! But I found it all to be stimulating; you gain a fairly deep knowledge of human pathology and medicine in that one year.

My internship ended in 1971, and at that time all graduating medical students in the country were drafted for the war in Vietnam. So, it was natural for people in my cohort to seek alternative service, such as at the NIH or the CDC or the FDA. I was interested in the NIH anyway as a place to do research, and so I applied to and was accepted into the Research Associate Program in Martin Gellert’s laboratory in the National Institute for Arthritis and Metabolic Disease, the forerunner of the current National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). There had been a history at NIH of supporting people with alternative service during the Korean War and now the Vietnam War—twenty years of what we later called, in contrast to the Green Berets, the “Yellow” Berets. The Yellow Berets were the forces who did their service at the NIH, and they reformed the NIH. There was this huge influx of the brightest medical students who were interested in biomedical research, so in a peculiar sort of way, the Vietnam War made all that possible. I stayed at NIH three years as a Research Associate and learned some sophisticated molecular biology from Marty Gellert, before going back to Peter Bent Brigham to finish my residency in medicine.

MI: So, you had to put off research for another year?

MG: Yes, but I was given credit by the American Board of Internal Medicine for my research time at the NIH, so I only needed an additional year of clinical work, and my advisers all encouraged me to complete the training, because it would give me more flexibility. And I did get boarded in internal medicine. During this next year, I was asked to stay on at Harvard as an Assistant Professor in the Department of Anatomy, which was a wonderful opportunity, but at the same time, my wife Susan Gottesman was at MIT working with David Botstein as a sort of senior postdoctoral fellow while she looked around for a faculty job. At that point, Susan and I were offered long-term positions in Ira Pastan’s laboratory of molecular biology in NCI. Having Ira’s support, and with the enthusiastic backing of Alan Rabson, the scientific director of our division in NCI, made a huge difference in helping us establish our independent research programs.

MI: As you began your nearly thirty years of research at the NIH, what did you start to work on?

MG: I had already established an independent laboratory at Harvard, but had been there just a year, and work was just getting off the ground, so I decided to continue with the research I had started at Harvard. Specifically, I was interested in factors that were secreted by tumor cells that might influence either blood vessel growth or metastasis. Judah Folkman’s work on blood vessel growth had been an incentive to some of the work in Bert Vallee’s laboratory when I had been there a few years prior. So, my first step, because of my training, was to characterize what was being secreted by malignant cells as opposed to nonmalignant cells.

MI: So you began your career here at the Cancer Institute?

MG: Yes, so the research was a good fit, and it seemed like a reasonable project. We began to characterize some of the factors that were secreted by tumor cells, one of which we called MEP, for major excreted protein—it was a very prominent band on acrylamide gels. We cloned it and demonstrated that it was a lysosomal protease, which later would be recognized as cathepsin L. It’s involved in a lot of interesting biological processes, but it doesn’t seem to be a major factor in cell transformation per se or for cell behavior in metastasis and invasiveness.

I also became interested at the same time in developing cell systems to do genetics in tissue culture. There was a burgeoning interest in being able to manipulate cultured cells genetically, to study all the genes that were involved in processes like growth and transformation. We started working on Chinese Hamster Ovary (CHO) cells, a system originally described by Ted Puck, which Lou Siminovitch had used to great fruition to identify mutants. Lou himself had a background as a bacteriphage geneticist, and he was interested in developing systems to study mammalian genetics. Lou sent me all kinds of strains and helped me get started. Nobody around NIH was really doing this kind of work back then.

MI: What kinds of problems were you hoping to address with these cell cultures?

MG: The process of cell division. Now remember, this was way back in the late 70s—we didn’t know anything about oncogenes. We didn’t know anything about what governed how cells divide, and we needed a way of genetically transforming those cells. We were among the first to transform cells with DNA to express specific characteristics in culture. The power of our approach was the combination of a system where we knew we could isolate mutants with one where we could transfer DNA. As cancer research has developed, such approaches have turned out to be critical.

And so we set out to isolate mutants that we thought might be affected in cell division. We knew that there were specific drugs, used to treat cancer, that affected the ability of cells to divide—drugs, namely, like the vinca alkaloids, vinblastine, vincristine, and although not used clinically, colchicine and colcemid. These drugs were known to bind to microtubules, so it was thought that they were interfering with spindle formation. And then another drug came along in the early 80s, taxol, which seemed to stabilize microtubules, but also interfered with spindle formation.

My first postdoctoral fellow, a fellow named Fernando (Buz) Cabral, had been working in Switzerland on developing 2-D gels to analyze mitochondrial proteins, and we thought this approach would be a good way to address the biochemistry and phenotype of our mutants. We were in fact able to show that some of the cells we had selected for in the presence of the microtubule-depolymerizing drugs carried mutations in alpha- or beta-tubulin, and we learned a lot about spindle formation. You can’t knock out spindles and have viable cells, but you can get conditional mutants that are either cold-sensitive or drug-dependent. Buz took this project with him when he went off on his own—he’s now a professor at the University of Texas at Houston.

MI: You didn’t mind letting someone carry an interesting project like that away from your laboratory?

MG: No, because another interesting project was also beginning at that time: Although some of our initial mutants were characterized primarily by resistance to the vinca alkaloids, others turned out to be cross-resistant to a wide variety of different drugs—not just to colchicine, for instance, but to all the anticancer drugs: doxorubicin, actinomycin D, and virtually any other chemotherapeutic agent that you might name. It was Bruce Chabner, who was the head of the Cancer Chemotherapy Division at NIH, who pointed out to me the phenomenon of multidrug resistance as a major reason why chemotherapy fails in patients. My colleague Victor Ling, at the same time, had begun to study a protein expressed on the surface of multidrug-resistant CHO cells, which he called P-glycoprotein—the “P” stood for “permeability.” He described the function of P-glycoprotein in terms of blocking cell permeability to these drugs, and it was at that point that we became interested in finding the actual molecular defect in our multidrug-resistant mutants.

Another bit of serendipity occurred at that time when I ran into a young scientist at a Gordon conference, Igor Roninson, who had developed a technique for cloning genes that were amplified in cells. Bob Schimke had already done his pioneering work showing that gene amplification was a common cause of drug resistance—he was working on resistance to antifolates, where he found amplification of the dihydrofolate reductase gene. So, Ira Pastan and I began to collaborate with Igor Roninson, using his technique to find amplified genes by virtue of their apparent relative facility to renature and appear as isolable bands on radiolabeled gels. By using that technique, we obtained a probe for what turned out to be the gene that was amplified in our multidrug-resistant cells and prepared a full-length cDNA, which, lo and behold, encoded the P-glycoprotein. In this way, we demonstrated that the same protein that was increased on the surface of multidrug-resistant cells was in fact encoded by an amplified gene. We also showed that our full-length cDNA could be introduced into sensitive cells that would consequently develop the full phenotype of multidrug resistance. We also demonstrated, based on the sequence, that the protein had two ATP binding sites, and that it worked as an energy-dependent efflux pump. Kinetic analyses showed that it recognizes the drugs in the membrane—so it actually does keep drugs out of the cytoplasm—and so one observes an initial little bit of uptake (into the membrane) followed by efflux.

MI: So, you know that P-glycoprotein is upregulated in multidrug-resistant cancer cells. What is it doing in normal cells in the first place?

MG: Ira Pastan and I collaborated to find that the protein was highly expressed in about 1000 different cancer cell lines, and we also wondered what it normally did. With Mark Willingham, a very talented histochemist and microscopic anatomist, we were able to show that P-glycoprotein occurs in the transporting epithelia of different tissues, such as in the kidney, in the GI tract, and in the liver, where it is situated appropriately as an efflux pump. And because a great many of the compounds that are used to treat cancer are natural products—they’re in foodstuffs, they’re in microorganisms—we concluded that P-glycoprotein provides a normal defense against toxic xenobiotics; evidence from knockout mice supports this conclusion. We have also made an MDR1 transgenic mouse, and we showed that the overexpression of P-glycoprotein in the bone marrow makes the mice much more resistant to anticancer drugs, which may have implications for gene therapy. The protein is also naturally expressed in the adrenal, which is a steroid-secreting tissue, and so we think that P-glycoprotein might have something to do with steroids getting out of adrenal cortical cells. It’s also expressed at barrier sites in the brain—capillaries in the brain—in the testis, and the ovary and in the placenta.

MI: So in all of those sites, P-glycoprotein is serving a protective function.

MG: Yes, and we know that polymorphisms in the gene for P-glycoprotein may affect the pharmacokinetics of many commonly used drugs, not only anticancer drugs. We’ve studied how the system works: Basically, hydrophobic compounds that cross cell membranes are recognized by certain sites within the P-glycoprotein molecule, where they activate the protein’s ATPase activity. There’s a conformational change in the protein that extrudes the hydrophobic compound out into the extracellular space. The process of pumping compounds back into the extracellular space is more efficient than the process by which the drugs can freely cross the membrane. The turnover number of P-glycoprotein is once per second, whereas free diffusion of substrate compounds into the cell occurs on the order of seconds. It’s a very clever mechanism! Nature is very clever.

MI: The science has also been clever!

MG: Yes, and I have two senior colleagues now. Suresh Ambudkar is a membrane biochemist who has worked with us, now independently, for the last ten years on the mechanism of action. And more recently we have a colleague, Di Xia,who is a crystallographer; we want to find out at the molecular level exactly how this protein works.

MI: P-glycoprotein was one of the first discovered proteins to be representative of a large protein family, wasn’t it?

MG: Yes, in the human. When we discovered the sequence of human P-glycoprotein, there were two known ATPase-dependent pumps in bacteria: the maltose transporter and the histidine transporter. But it wasn’t appreciated until all three sequences were compared—and also the sequence of the mouse P-glycoprotein that Philippe Gros published about the same time—that the energy-dependent transporter systems would prove to be so prevalent. There are forty-eight in the human alone; three are known to be involved in drug resistance in cancer, and a dozen can confer drug resistance. We’ve been lately using a more genome-wide approach to catalog the roles of the forty-eight transporters—we’re using microarray and RT-PCR approaches and looking at many, many different cancers. The story is incredibly complicated—and this is only a small part of the drug resistance story! There are many other ways that cells can become drug-resistant. And the take-away lesson for cancer research is, as complex as it is to understand cancer in terms of oncogenes, recessive genes, and dominant genes, it’s equally complex to understand the ways in which cancer cells can defeat our efforts to treat them with anticancer drugs. And we need to understand both aspects. We need to have the targets of the drugs and we need to have an understanding of how cells handle and become resistant to those drugs. If the most elegantly designed drug doesn’t get into the cell because of these very commonly expressed transport systems, that drug isn’t going to do you any good at all.

MI: In addition to your own research activities, you’re also Deputy Director for Intramural Research at the NIH. What made you want to take on that role?

MG: That was also a bit of serendipity. Bernadine Healy, Director of NIH in the early 90s, and I had been classmates in medical school, and so when Jim Watson was stepping down as Director of the Genome Project, she was trying to find someone who could carry on. It seemed like an exciting challenge, so I spent a year as Acting Director of the Genome Research while we searched for the Project’s next Director, which turned out to be Francis Collins. And that was a very interesting year: I learned about managing a large organization and dealing with people’s frustrations and their excitement. It was a very talented group of people, so it didn’t require a lot of special effort on my part. And then with Francis Collins as Director, I was going back to the Laboratory when Harold Varmus became Director of the NIH and tapped me to be Deputy Director for Intramural Research, which was a job much more in keeping with my experience. I had been in the intramural program for many years. I understood and really loved the intramural program and the scientists in it, and I saw this as an opportunity not only to help foster the science, but to improve the training activities and opportunities on the campus of the NIH. And so I took that job ten years ago—ten years and three months ago, but who’s counting?

MI: Talk about the training opportunities that you’ve helped to develop at NIH.

MG: We started by reconfiguring the intramural program, implementing a much more rigorous oversight system of research for our tenure-track systems. We have a high school program that I helped to develop, which brings in high school students to work in the laboratories. We’ve established the Undergraduate Scholarship Program, aimed at students from disadvantaged backgrounds, in which we pay their tuition and expenses at college for a year and provide a summer research experience, and then for every year that a student gets support—and it can actually be up to four years—the NIH is owed a year of service later on as a postdoctoral fellow or a postbaccalaureate fellow. We’ve now had a couple of hundred students in this program—fabulous students, all of whom have gone on to bigger and better things. At the medical school level, there had already been a program at the NIH supported by the Howard Hughes Medical Institute, so that students could do medical research for a year. But on top of that, we developed a clinical research training program, now supported by Pfizer, which allows medical students to come here after their third year to do clinical research, and this has also been a very successful program.

MI: So, you’ve become an educator in the traditional, academic sense.

MG: Oh yes, that’s great fun for me, interacting with the students at all levels. With the enthusiastic encouragement of our NIH Director Elias Zerhouni, we’re working now—our most amazing effort of all—to revamp postdoctoral training in an interdisciplinary sense. Postdocs, frequently, in many institutions, are the lost, forgotten trainees—and they really should be regarded as trainees. They’re beyond their PhD years, but they’re interested in learning something that will carry them through their careers. And so, we’re trying to develop much more formal programs to make sure that their laboratory experiences are good, that their mentoring is excellent, and that they can be trained to do something that is really going to be important. With the enthusiastic support of Harold Varmus, we also developed a partnership program for graduate studies with the NIH. We’re not a university—we can’t give degrees—but we now have 300 graduate students on campus, working in our laboratories, who will be awarded their degrees from different universities.

MI: Beyond providing training to students and postdocs, what other purpose does Intramural Research at NIH serve?

MG: Intramural Research started out as the only NIH program in the 1940s, and although the extramural grant program has undergone enormous rowth, Intramural Research at NIH remains distinctive in several respects. Because of our relatively stable, long-term support, intramural scientists can undertake high-risk, complex research that is difficult to do with extramural grant support. For example, because of the clinical center, a 240-bed research hospital, we are uniquely positioned in translational and clinical research. We have one of the major academic vaccine development programs. We have at the NIH a vaccine research center whose main goal is to develop vaccines against HIV, and they’ve also had some success with Ebola. They already have some vaccine candidates that have begun clinical trials. My office administers an intramural AIDS-targeted antiviral program, which uses the physical and biophysical skills of NIH scientists to study the protein components of HIV—to solve the structures that may become targets for new drugs. The integrase structure, for example, was done at NIH. Various aspects of imaging, genomics, and informatics are also very vigorous in Intramural Research. We have a huge informatics program—the National Library of Medicine has over 400 people doing bioinformatics. We want the intramural program to be complementary to the work that’s going on at universities and research institutions and in the private sector, and we think we have a distinctive role.

| Table of Contents

Navigate This Article

  1. Top