Interview: Mary-Claire King

Mary-Claire King is best known for her research on the genetics of breast cancer, a disease that strikes one out of every ten American women in her lifetime. In this work, and in her research on inherited deafness and inherited susceptibility to HIV infection (among other subjects), Dr. King has demonstrated the power of approaching biological questions from a variety of perspectives. Depending on the question at hand, she may focus on molecules, on cells, on human families, or on whole populations-using whatever techniques are most appropriate. In addition to basic research, King's laboratory group applies the concepts and methods of genetics to solving practical problems relating to human-rights abuses around the world. Jane Reece spoke with Dr. King on these and other topics at the University of Washington, where she is a professor in the Departments of Medicine and Genetics.

How did you get started in biology?

After majoring in math at Carleton College in Minnesota in the mid '60s, I went out to UC Berkeley to study statistics. I wanted to try to integrate math, which was something I liked and was reasonably good at, with some sort of work that mattered to people. In those days we were very much involved with the civil rights movement, and we were becoming increasingly involved in what was going on in Vietnam. It wasn't an era in which it made sense to ignore the world around you. My first quarter at Berkeley, I took Curt Stern's genetics course, just before he retired. I remember going to that class every day at 1 o'clock, and it was like Christmas every day. I couldn't imagine that people got paid to work on these fascinating kinds of problems! That course changed my life. Soon I transferred to Genetics.

What kind of research did you do as a graduate student?

I worked in Allan Wilson's lab on questions relating to human evolution. With little background in biology, I was initially a menace in the lab, but I didn't let that stop me. My dissertation asked, "How similar at the level of genes and proteins are people and chimpanzees?" We discovered, much to our amazement, that they are extraordinarily similar, that we differ in only about one percent of our DNA. Morphologically and behaviorally, humans and chimpanzees seem to be in different taxonomic families. But genetically we are closely related species. How does one resolve this paradox? One possibility is that similar genes in humans and chimps are turned on and off at different times during embryonic development. The idea that the evolutionary divergence of species might be at least partially explained by changes in the timing of gene activity is one that other researchers are exploring.

What led you to study breast cancer?

I finished my Ph. D. in '73 and then went to teach at the University of Chile in a Ford Foundation program. I was in Chile when the Allende government was overthrown. There had been wonderful opportunities for the integration of scientific and economic development of the country, all of which were destroyed in the coup. I came back to Berkeley at a loss, thinking, "What shall I do next?" I knew that I liked genetics very much, and I also knew that I needed to do work that was important to people in an everyday sort of way. The first thing that came up was the possibility of working at UC San Francisco setting up a modern genetics lab in their cancer epidemiology group. At that time a lot of money was available for bringing approaches from other fields into cancer research. I was hired to set up a cancer genetics lab and was captivated by the question of what causes cancer, in particular breast cancer.

When scientists say that cancer is a genetic disease, what does that mean?

It means two things. Cancer is always genetic in the sense that cancer is always the consequence of changes in DNA. Cells that have cancer-causing mutations no longer divide and develop in the way that they should. Something goes wrong with cell division; it's no longer under normal control. The great majority of mutations that lead to cancer arise in the tissue where the cancer starts. The first occurrence is in the colon or in the breast, for example. These mutations are what we call somatic mutations; they are in the body but not in the germ line, the cells that give rise to eggs or sperm. But very rarely, in some families, there are germ line mutations in one or more of these same genes. These mutations are passed on from parent to child.

So breast cancer is not usually inherited?

The vast majority of breast-cancer cases seem to have nothing to do with inherited mutations. However, there are many accounts, going back to the ancient Greeks, of families in which breast cancer appears frequently, generation after generation. Of course, this doesn't prove that the disease is inherited. Lung cancer, for example, also clusters in families—but mainly because smoking clusters in families. So, does breast cancer cluster in families because of their exposure to some unidentified environmental agent, or because of an inherited genetic trait, or some combination of both? I thought it would be possible to sort that out by combining statistical studies of families with experimental genetics, and it was. But it took 20 years.

Did you first approach the problem, then, by collecting data on families in which breast cancer was common?

Yes. It occurred to me in 1974 that if mutations were involved in familial breast cancer, we might be able to identify some of the genes that are mutated and then figure out how these genes normally work. The results might give us insight into the more common, nonhereditary forms of the disease.

You eventually identified the gene BRCA1, which is mutated in many families with inherited breast cancer, and located the gene on chromosome 17. Are mutations in BRCA1 also found in the tumors of women with nonfamilial breast cancer?

If you test tumor cells from women with nonfamilial breast cancer, you find that cells from more than half of the tumors completely lack one copy of the BRCA1 gene (of the two in each normal cell) and make reduced amounts of BRCA1's protein product. But the sorts of breast-cancer mutations inherited in families are not found in tumors from women with nonfamilial breast cancer. This paradox still has not been resolved.

How do you sort out hereditary influences on breast cancer from environmental influences?

We're now addressing that question in a project with Jewish families in New York City and Israel, families with known mutations in BRCA1 or BRCA2, a gene found soon after BRCA1. (Among breast cancer patients of Jewish ancestry, about 10% have one of these mutations.) With the participants' permission, we test the DNA of breast cancer patients from these families for mutations in BRCA1 and BRCA2. For each patient who carries one of these mutations, we trace the history of the mutation in her family back as far as possible. Then we ask, "Did all of the women in the family who inherited the mutation develop breast cancer? At what age did cancer appear?" If some women developed breast cancer at 70 and some at 30, were there any differences in their environmental exposures? If some women with the mutation lived to a very old age and never developed breast cancer at all, can we explain that?

What is your view of genetic testing?

A woman carrying an inherited mutation in BRCA1 or BRCA2 seems to have a very high risk of breast cancer (and ovarian cancer as well), a more than 80% risk of developing cancer during her lifetime. But does testing women for mutations in these genes make sense? To me there are three principles that are important to consider. One is that testing must be accompanied by patient education. Testing is useful only if it is presented in such a way that the person understands what the limitations of the test are and what the results mean. Genetic counselors specialize in conveying that information, and their participation is very helpful.

The second critical principle is one of social justice: The social context in which genetic testing takes place needs to respect absolutely the rights of the individual. A person's genetic background should have no bearing at all on his or her ability to obtain health insurance, for instance. It seems to me a very simple concept. We are all predisposed to something. Although researchers have sorted out only a relatively small number of disease-predisposing genes to date, the number is growing daily. If health insurance depended on all genes being "normal," there wouldn't be any insurable people! So, as a matter of both logic and justice, it makes sense to separate the availability of health insurance from a person's genetic makeup.

The third principle has to do with the pace of research in biology. For a woman today to know that she is predisposed to breast and ovarian cancer offers her a problem-but no solution except preventive surgical removal of her breasts and/or ovaries. I wish research moved so quickly that disease-predisposing genes could be identified and treatments developed within months of each other, but that is clearly not the case. The development of a useful intervention is enormously important but is also usually enormously difficult. So we need to work out a rational policy for genetic testing that recognizes the need for education and justice, and the realities of biological research.

Tell us about your lab's involvement in projects relating to human rights abuses.

Our first project, which continues now some 15 years later, was the identification of children who were kidnapped as infants during the Argentine military dictatorship between 1975 and 1983. The children's parents "disappeared." When the grandmothers of these children realized that kidnapping had been widely used as a political tool during that period, they organized and began a search for the missing children. Because the children would not know who they were, the grandmothers needed a geneticist. In 1984, I became that geneticist.

We originally used classical markers for the identifications-blood groups and HLA types (the protein types used to match tissues for organ transplants). But within a few years we began to use newer methods based on DNA technology, such as mitochondrial-DNA sequencing (determining the order of nucleotides in the small DNA molecules of mitochondria). Mitochondrial DNA is ideal for the purpose because it is inherited only through the mother and is highly variable from family to family.

The grandmothers know of about 220 children who were kidnapped, of whom 59 have been identified so far. We have a database of DNA sequences from families who lost children and grandchildren. Now, when young adults come to the grandmothers suspecting they may have been kidnapped, we can compare their mitochondrial DNA with the sequences in our database. We can often tell these young people who they are and put them back in touch with their biological families.

That work evolved in the early 1990s to the analysis of DNA samples from murder victims. Our first cases were some of the murdered mothers of the kidnapped children. And we've since extended our DNA-identification activities to victims in other countries. We've helped identify the remains of MIAs from the Vietnam War, for example, and now we're working as the Molecular Genetics Identification Laboratory for the UN War Crimes Tribunal. We're identifying victims of human rights abuses in Bosnia, Croatia, Rwanda, Somalia, Ethiopia, Chiapas, Guatemala, El Salvador, and Colombia. So we have many projects under way.

What advice would you offer an undergraduate biology student who is interested in a career in genetics?

The field is moving so quickly that the opportunities to answer fascinating, important questions are without limit. The work itself requires devotion, but the effort is easy when you are enchanted by the subject. Do not be discouraged if the first two or three or four things you try aren't a perfect match for your interests and abilities. Have the courage to go up to a professor and say, "The research going on in your lab intrigues me, and I would like to understand it better." It takes a lot of initiative to do that, but it is worth it. One really does need to move beyond courses and have real-life experience doing biology in a laboratory or field setting in order to get a sense of what it is about. In my own lab of 21 people, six are undergraduates.

Do the undergraduates get to make creative contributions to the research?

Oh yes. They come up with ideas, do experiments, publish papers, and go to scientific meetings. But no one does these things the first day! To do research, you have to master a set of skills, as with any other new activity. It's like learning sewing or woodworking. It takes a while to get to the point where you can be creative.

©2005 Pearson Education, Inc., publishing as Benjamin Cummings