A fever fueled Stanley Fields’ invention of the two-hybrid system for detecting protein interactions. Happily, his passion for devising new ways to study biology’s messy problems still burns hot.
Stan Fields was in need of funding. As an assistant professor at the State University of New York, Stony Brook, in the late 1980s, he was studying a transcription factor involved in yeast mating and pheromone response. And like many young faculty members, he had an eye out for new grant opportunities. That’s when the university’s Center for Biotechnology announced that it was offering seed money for projects with commercial potential. “It was $35,000—I remember that quite vividly,” laughs Fields. “That seemed like a lot of money back then.”
So Fields put on his thinking cap. “I...
“It was a Friday afternoon and I was feeling kind of feverish—I think I was coming down with the flu,” he says. “Then this idea just came to me: Instead of having a binding domain hooked directly to an activation domain, what if you had a binding domain hooked to one protein, and an activation domain hooked to another protein?” If those two proteins interacted with each other, they would bring together the two halves of a hybrid transcription factor. That transcription factor could then switch on a reporter gene such as β-galactosidase and, voilà, the system would give investigators a nifty way to monitor protein interactions. Thus was born the idea for the famed “two-hybrid system.”
After his fever broke, Fields quickly wrote up the application—and quickly received a rejection letter. “I called the director of the center,” says Fields. “He said they didn’t see the commercial potential.” But Fields was undeterred. He set to work on developing the technique and soon after scored $150,000 from Procter & Gamble to fund his exploratory efforts.
The resulting two-hybrid system quickly became the predominant means for tracking who’s touching whom inside the cell—and Fields and his colleagues continued to adapt the technique for detecting RNA-protein interactions, as well as large-scale networks of protein interactions.
Here he waxes philosophical about the difference between biology and methodology, the death of paper journals, and his desire to pen a lab-based murder mystery.
Up in flames. As an undergraduate in the late 1970s, Fields landed a two-year fellowship at Trinity College, Cambridge. He wanted to stay to do his graduate work at the nearby MRC Laboratory for Molecular Biology. “But I needed a letter of recommendation from Hans Kornberg, a very distinguished bacterial geneticist who was teaching one of the laboratory courses,” he says. “I was in the lab plating out E. coli with a glass triangle you’d flame to kill the bacteria between one plate and the next. I had dipped my glass triangle in a beaker of ethanol and passed it through the Bunsen burner flame when a drop ran down the rod and onto my bench. So I had this little bit of flame on my bench. As I went to take care of it, my arm knocked over the beaker of ethanol and my entire bench went up in flames. Hans Kornberg walked over, puckered up, went, ‘phwooo phwoo phwooo,’ and blew it out. And I thought, ‘I’m not going to ask him for a recommendation today.’ ”
Just be you. “In the first couple of years of my faculty career, I was still doing more or less the stuff I had done as a postdoc: figuring out a signal-transduction pathway in yeast that led up to transcription. We were making some headway, but I wasn’t that excited by what we were doing. It wasn’t until I started working on two-hybrid stuff that all of a sudden I said, ‘I like doing this a lot more than I like doing biology-driven research.’ And I stayed with that because it fit me. I think if I had continued working on biological problems—trying to figure out how things work in the cell—I would not have been as happy and I doubt I would be as successful.”
Biology’s a mess. “When you work on biology, there’s always a right answer—more or less. But methodology is completely unconstrained. You can come in to the lab, sit down, and say, ‘How about if we develop a method to do such and such?’ And if it works, you get a very clean answer: the method does what you set out to do. Biology is always messy and often very complicated. Methodology, for the most part, is very simple, once you have the idea. It either works or it doesn’t.”
No wall necessary.“For many scientists who work on methods, the motivation is to develop a method; it is not to solve a biological problem. Most people have this notion that you develop a method because you’re trying to solve a biological mystery and you’ve hit a wall. But a lot of methods don’t come about that way. They come about because someone like me says, ‘Here’s an idea. Let’s figure out how to do it.’ ”
A method called Sally.“I didn’t have a name for two-hybrid. The original paper called it, ‘a novel genetic system to detect protein-protein interactions.’ Then I was talking with my old postdoc advisor, Ira Herskowitz, telling him about the two-hybrid experiments. And he said, ‘So, Stan, what do you call this method?’ I said something like, ‘detection of protein-protein interactions by interaction of a Gal4-DNA binding domain fusion with a Gal4-activation domain fusion.’ While I’m saying this, Ira is just shaking his head. He appreciated that if you want to explain something to people, you have to have a simple name. I went back to the lab and we came up with ‘two-hybrid system.’ ”
Stop the clock. “It’s clear that women would be more likely to succeed in science if they had a little help juggling their family life and professional life. As a community we still have not come close to sorting this out. Here’s an idea that wouldn’t cost any money: We all have these grants where you put your salary on the grant. A typical NIH RO1 grant lasts for four years and the clock starts on day one, which makes it very hard to take any kind of leave. Why not adopt a system where you start with a certain percent of your effort on the grant—say 50 percent—and you can adjust that figure month by month? If you have a kid and you’re out of the lab for a month, you can be at zero percent effort. Then you can ease back in: a few months at 10 percent, then 20 percent, and eventually you’re back to 50 percent. If you started with a four-year grant, it might then become a five-year grant. And you wouldn’t even have to think about it. I pitched this idea to an NIH director once, but it didn’t go anywhere.”
What’s mine is mine. “I was disappointed when I came back to the United States from my graduate work to find out how stratified the American system is. Everybody owns their own stuff. At the MRC, everybody shared equipment and reagents, and you could get glassware from anywhere. It was a very communal system that operated very efficiently. It’s unfortunate that the American granting system doesn’t support a little bit more of that kind of environment.”
Walk on the wild side. “I tell graduate students to try wild things—try ideas that might be off-the-wall. This is the time in your life to have fun, be creative, tackle tough projects, and chase big ideas. If you become cautious right at the beginning, I think science becomes too much like a real job as opposed to something we do because we enjoy it and we’re passionate about it.”
Fund ’em while they’re young. “We need to get more people funded early in their career. I would put more money into RO1 support and not as much into center grants and big projects. Those big projects work, but they don’t necessarily generate the same kinds of exciting data that individual investigators do. I would rather place bets on a lot of people at a lot of places—give them the money and see what happens.”
In situ peer review. “At the kind of journals we all want to publish in, authors are mostly dealing with professional editors: people who’ve done graduate and postdoctoral work but have never run their own labs. It’d be nice to see more practicing scientists on staff at these high-profile journals. Then you could deal with someone who would be able to look at reviewer comments and say, ‘I agree, this suggestion doesn’t make sense and you shouldn’t have to do it.’ I wonder if there couldn’t be a sabbatical program where you’d go to work at Nature or Science or some other journal to learn about publishing for six months—like how NSF hires people for a year’s sabbatical to learn the administrative side of grant managing. If you had enough people doing these sabbaticals, you’d always have working scientists at the top-tier journals.”
Minding your manners. “I think it’s important, as a scientist, to have a certain kindness. Our competitors are often doing good science, worthy of recognition—and sometimes we don’t acknowledge that. And we should appreciate our students and postdocs, even when they don’t give us the results we’re looking for. Sure, there are successful scientists who have trampled on all kinds of people—but they’re not role models. We all have a certain obligation not to behave like complete jerks. Not all of us manage to do that all the time, but I think the world of science would be better if we did.”
E-liminating ideas? “Ideas are ultimately built on other ideas. Often it’s the odd bit of knowledge from some other field, some odd conversation you had, or some seminar you attended that gives you an idea or gets you to the next stage.” The move away from paper journals threatens that sort of synthesis. “Nobody browses journals the way they used to, when you’d happen on papers and read the abstract or glance at the figures. That’s a real loss.”
Scientific hero: Fred Sanger, in whose division Fields completed his PhD. “This guy has had more amazing ideas about technologies than anybody I’ve ever seen or heard of. And I think his philosophy has been, ‘Let’s develop the methods and then there’ll be all kinds of interesting biology that comes out of it.’ When he finished a method, he didn’t pursue the biology; he went on to develop another method. If I’ve had any big hero in science, it’s definitely been Fred.”
It’s a vocation. “The great thing about my life in science is that my kids have grown up seeing me very satisfied in what I do: working much more than “9 to 5” and at times becoming really obsessed with figuring things out. That’s part of the passion that we have. And I think that’s a good thing for your kids to see.”
Reaching out. With Mark Johnston, Fields has written a book, called Genetic Twists of Fate, designed to explain modern genetics to a general audience. “We’ve taken simple genetic principles and tried to wrap them in interesting stories about people who’ve had a genetic disease or scientists who’ve worked on the problem.” His biggest fear: that the book will sink without a trace. “If the book is a failure, it’s a failure. But I’d rather it be a failure because people look at it and say, ‘There’s nothing in here we like,’ than have people say, ‘What book? I never heard of it.’ ”
Whodunit, PhD. “I’d like to write a novel—a murder mystery where the reader ends up inadvertently learning some biology along the way. I have the whole plot worked out. A graduate student is killed, and he’s actually not a nice guy. Potential suspects are his girlfriend, his advisor, and some other guy in the department. The book would include a lot about evolution, and what it’s like to be a professor trying to keep a lab going. I’d definitely like to write for a general audience and convey the excitement of science to a wider group.”
Rewind. “At some point before I end my career I would like to relive history and be a grad student again, without the pressure to get a degree or to get a job. Just go back and tinker.”
As a graduate student working with postdoc Greg Winter, sequenced an influenza virus genome—work that yielded more than half a dozen papers, including two in Nature and one in Cell.
Developed the two-hybrid system for detecting protein-protein interactions in vivo; with Marvin Wickens, adapted the technique to detect RNA-protein interactions.
Synthesized a library of 6,000 yeast proteins fused to the Gal4 activation domain; used these fusion proteins to conduct genome-wide, two-hybrid screens.
With Eric Phizicky, developed a method that takes a biochemical activity and rapidly identifies the protein—and the gene—responsible.
Assembled earliest diagrams of the complete network of protein interactions in yeast—”the first of the infamous hairballs,” says Fields.
Carried out large-scale characterization of the effects of mutation on protein function.