Choosing the T-Cell Profession: Higher Education for Stem Cells

Your body is continuously making new blood cells from a reservoir of "starter" cells called stem cells. Blood cells come in many types, including the highly versatile T cells that play a number of key roles in the immune system. All stem cells are alike, and all the T cells that come from them start out alike before choosing specific careers in response to signals from their environment.

On Wednesday, November 18 at 8 p.m. in Caltech's Beckman Auditorium, Ellen Rothenberg, Caltech's Albert Billings Ruddock Professor of Biology, will lead us along the paths that T cells follow and show how her lab has mapped their journeys. Admission is free.

What do you do?

I'm interested in how cells choose their identities through reading out information stored in the genome, which is the entire collection of DNA that makes a creature what it is, and how a cell that begins with one identity can spawn descendants with very different, very durable new identities.

We study T cells, a large family of white blood cells that form a major part of your immune system. T cells have an extremely long and varied life. They come from so-called stem cells, which have the ability to become many, many different kinds of cells. We want to learn how a "blank slate" of a stem cell develops to achieve a rock-solid identity as a T cell—especially because a T cell has an irreversibly defined "T-cell-ness" at its core, yet it remains very dynamic in using genomic information to decide what kind of T cell it will be.

Generating T cells is a three-step process. First, a stem cell develops into a T cell. Second, the T cell circulates around the body, waiting to see how it will first be used by the body to fight an actual infection. And then third, once it has evolved a specialization, it will continue to go around the body for months, years, or even decades in humans, spawning descendants that are also specialized with the same specific type of cellular function the original T cell had when it was activated—as helper T cells, or killer T cells, or whatever other type of T cell was needed. And they may pick a subspecialty—for example, for every infectious agent you encounter, you develop a specific memory cell to recognize that particular bug so that if it comes around again you are ready for it.

Once made, the decisions are locked in. All the T cell's progeny will generally stay in "the family business." However, it sometimes happens that once a T cell has chosen its profession, a particularly strong environmental signal can drive it to change into a different type of T cell. But even so, it will never, ever go back to being a stem cell. My lab is trying to figure out the molecular control mechanisms that allow the former stem cell to achieve a new rock-solid identity as a T cell, yet maintain a level of flexibility within that T-cell-ness.

Why is this important?

I study biology for the same reason that astronomers study the universe. I believe that there are deep biological principles to be learned from T cells, whose import goes way beyond curing a particular disease. I'm ecstatic when things we do are picked up by clinicians, who do make a profession of helping people, but I do basic science.

There are two main branches to the developmental biology of multicellular organisms. The first goes from the fertilized egg through the embryo, and that's the process that makes your body in the first place. It follows well-known rules worked out by people like my late colleague Eric Davidson [Caltech's Norman Chandler Professor of Cell Biology].

I study a second form of development that begins when an embryo sets aside a bunch of cells and programs them to become stem cells. Stem cells do not differentiate further right away; they just make more copies of themselves. Then, whenever you need to make new blood cells or repair a tissue later in life, those cells are called into action. For example, red blood cells only last about three or four months, so the blood circulating in your body today is coming from stem cells, and those stem cells were "set aside" when you were a fetus. This means there's an additional set of rules, going well beyond embryonic development, for making new blood cells in the right balance and at the right time.

The new cells do have some wear and tear from the consequences of your adventures throughout your life, but to a first approximation they're the same. They're getting primed to do the same job. They have to set up all the molecular circuitry needed to retain their identity and maintain a clear one-directional flow from stem-ness to differentiation. The process has to be as accurate at our advanced ages as it was when we were fetuses. That's the genius of stem-cell-based developmental biology. In my view, the collection of stem-cell development mechanisms ranks right up there with the more established mechanisms of embryonic development.

How did you get into this line of work?

I've always been interested in science. The question when I was young was whether I wanted to be a physicist or a biologist, but then I fell completely in love with biochemistry when I was in high school. When I went off to Harvard I didn't know specifically what I was interested in, but I loved what was known about the genome. I thought it would be fantastic to understand how the genome works at a molecular, mechanistic level.

I had the great good fortune to have microbiologist Boris Magasanik as my undergraduate tutor and mentor. He was the head of MIT's biology department, but he had a relationship with Harvard and he liked teaching undergrads. Boris was an extraordinary intellectual. He was studying metabolic pathways in bacteria at the systems-biology level way before it was normal. He was drawing prototype diagrams of gene-regulatory networks back in the early '70s.

A lot of technology had to be invented before we could explain gene regulation on the molecular level, but when I became a graduate student in [Nobel Laureate] David Baltimore's lab at MIT in 1972, he was already doing incredible work on viral genomes. [Baltimore came to Caltech in 1997 and is currently the Robert Andrews Millikan Professor of Biology.] We were pushing the frontiers of knowledge outward on a daily basis, and it was exceptionally exciting.

However, the development of multicelled organisms was still extremely hard to understand back then. It seemed all anecdotal, as if every organism did things in a fundamentally different way. But by the late '70s, Eric Davidson here at Caltech was making it possible to make sense out of developmental systems. His views integrated Boris Magasanik's systems-level view with David Baltimore's molecular-level finesse, and his work was revealing general mechanisms of development in multicellular organisms. I owe a great deal to the conceptual and mechanistic perspectives that I have gotten from these three people.

Also, Caltech's smallness has been fantastic. Most of the people I know who work with T cells are in immunology departments, and most immunologists do the same kinds of things, more or less. The joy for me at Caltech has been doing things that nobody else is doing. Often when my colleagues here solve their problems, I can use those approaches to break new ground in my field. It's been extraordinarily fun, and a tremendous advantage. Science as it should be done.

Long ago at MIT, my labmates and I were studying a retrovirus that caused early T-cell leukemia in mice. Lots of retroviruses cause cancer by putting a gene responsible for normal cell growth into the host cell and then turning the gene on under the wrong conditions. But our retrovirus didn't cause cancer in other cell types, so we wondered why it affected early T cells. I realized that the T-cell development process itself must be an especially sensitive target. The retrovirus nudged the future T cells toward being cancerous, possibly by accident, and then a little push farther down the line would send them over the edge.

That's when I became interested in T-cell development and this question of what controlled the switchover between growth and differentiation. We've found in the last 10 years or so that there are actually two bursts of proliferation during T-cell development. My lab has focused on the first one, which we now know is the transition between stem-cell-ness and T-cell-ness, when the cell commits to becoming a T cell. And it turns out that if a stem-cell regulatory gene stays on during the process, you get an abnormal persistence of stem-cell-like growth and sometimes leukemia. It's ironic that it's taken me, gosh, 40 years to get back to that, but it has been an incredibly satisfying journey.