Gaudenz Danuser: The symphony of the cell

You seem to take a very analytical approach to your work…

Well, my father is an engineer, so I probably have that in my genes. But my mother was a music teacher, and actually, before I became a scientist, I was myself a musician.

What instrument did you play?

I started with the trumpet, which became my main instrument at the professional level. I also played the piano because it’s mandatory to learn it as part of the musician’s professional track in the music academy. I picked that up at the age of 14 or 15, so the piano and trumpet were my two main instruments. I also composed some of my own music; I was looking back through my sheets the other day and saw that I wrote my first piece at age 13.

Do you still play or write music?

No, I do neither. The problem is, I used to play at a professional level, and it would take me three to four hours a day of practice to just maintain that level. An academic career is quite time consuming, so that simply isn’t possible now. I hope in the future—maybe after I retire—to find time for that in my life again. For now, though, I am just a very active consumer of music.

So what led you to a research career?

During high school in Switzerland, where I grew up, you can enroll in the normal high school program and also in some after-school and lunch programs to attend the music academy in parallel. That is what I was doing, but at some point I began to worry: is music a bread-winning career, given the enormous competition I would face? So a year after graduating high school I enrolled in the engineering program at ETH Zurich. That was a four-year program, and I expected that afterwards I would get some kind of engineering job to support myself while I pursued my musical interests.

But things turned out differently. When I graduated, I spent a year in industry and then returned to ETH Zurich to complete a PhD, also in electrical engineering. During my PhD, I worked on microscopy—but for robots, not for biology. There was a campus-wide program to create a robot system that could manipulate micrometer-sized objects with nanometer precision, and I was working on its eye, which was a microscope. At the end of my PhD program, I realized that the people who could really use measurements through microscopy are biologists. So I began looking for postdocs in biology, and that is how I came to work at the Marine Biological Labs in Woods Hole, with Rudolf Oldenbourg and Shinya Inoué.

Your work there seemed to undergo a big revolution…

Yes. [Laughs] At Woods Hole I worked with polarization light microscopes to make very detailed maps of actin filament flows in cells. The discovery I made as a postdoctoral fellow was that these flows are differentially regulated in space; there might be more flow on the left side of a cell than on the right side, and these differential flow patterns are directly coupled to the shape of the cell. I was writing up that paper when Ted Salmon visited for the summer. He was a former student of Shinya’s, so of course he was hanging out in our lab, and he saw what I was doing and immediately invited me down the hall to talk to Clare Waterman-Storer, who was labeling actin networks with rare fluorophore speckles. I was totally blown away by her speckle images because the resolution of that technology was so much better than what I was working with. I realized that we could easily use it to get the kind of information I had been working to get for the past two years. On the other hand, I recognized that this was going to be an analytically difficult problem, so I immediately established a collaboration with them to analyze these data. It worked out fabulously.

What was so difficult about this?

The big barrier we faced was to understand what a speckle actually meant. It took us a few years to realize that a speckle is really a relative signal. A speckle can appear in an image either because fluorophores are added locally or because they are removed from the surroundings. That means they can both appear and disappear during filament polymerization. Additionally, when depolymerization of filament networks is also taking place, new speckles could appear because fluorophores were stochastically removed from the local background but not from the foreground. When we understood that and adjusted our software to account for it, then it became possible to directly measure polymer assembly and disassembly in living cells. I would say that was the big eureka moment for us with that technology.

This was after your postdoc, while you were at Scripps Research Institute?

That’s right. But I didn’t go directly to Scripps after my postdoc ended. I first returned to Switzerland for four years, as a lecturer and then assistant professor at ETH Zurich in the mechanical engineering department. That was a difficult time for me because, at that time, Zurich wasn’t ready for the kind of interdisciplinary, quantitative biology that I wanted to do, so most of my collaborations were back in the United States. It was so frustrating that I almost left academia. I was about to sign a contract with Roche when I got a call from Sandy Schmid at Scripps, who probably had an eye on me because of my visits there to work with Clare. She asked me what it would take to bring me over there, and I said, “Probably not that much.” That was my first—and last—negotiation mistake. [Laughs]

You work on many subjects besides the cytoskeleton and cell migration…

I think our willingness to collaborate is a big part of the success we’ve had. As an engineer, once I have the technology to solve one problem, I really try to think about generalization. Many of these analytical tools—particularly those for image analysis—are built for one purpose initially, but they can be used for other problems as well. So I very much enjoy being approached by people with new questions and have several great collaborations going on various topics. For example, Sandy Schmid approached me with questions about endocytosis. One reason I like working on endocytosis is that it is a very accessible process for studying heterogeneity—not just between cells but between different molecular assemblies within one cell. Once we can understand and model this heterogeneity, we can generalize it to many other biological phenomena. We can also make tiny, targeted perturbations to study how different components contribute to the whole process.

What have you been using speckle technology for lately?

It can be used to examine any macromolecular assembly. For example, besides actin filament networks, we’ve also used it to characterize the dynamics of focal adhesions and the mitotic spindle. But one of my favorite pieces of work from my lab was from 2008, where we took this speckle technology to the next level and used it to look at the deformation of polymer networks. Once you can measure deformation, you can ask: How much force do you need to deform the polymer networks? When polymer network flow patterns change, what changes in force are occurring?

We developed new mathematical algorithms to answer exactly these questions, and we were able to dynamically map out forces within the cell. Now we can link these forces to signaling and begin to reconstruct the symphony of cytoskeleton regulation that is involved in cell migration.

Where are you headed in the future?

A large number of people in my lab are working on whether we can take these imaging experiments, look at the fluctuations, and mathematically tease out functional causalities between molecular events. Now we have a lot of good tools to help us closely examine these fluctuations. Perhaps I’ve been lucky to have picked the right problem at the right moment.