Naturally occurring enzymes are fantastic catalysts, but they are limited to reactions important in living systems. As a result, researchers have been trying for years to create 'designer' enzymes to speed up all manner of reactions under a host of conditions. “Designing enzymes from scratch has been a holy grail of computational structural biology,” says David Baker, a biochemist at the University of Washington in Seattle. He and his team recently succeeded in engineering enzymes to catalyse not one but two reactions for which no naturally occurring enzymes exist (see page 190 and L. Jiang et al. Science 319, 1387–1391; 2008).

In the work reported in this issue, the researchers set their sights on the Kemp elimination — a well-characterized reaction for proton transfer from carbon. They produced eight synthetic enzymes that could catalyse this process. Baker credits his team's success to an exceptional group of graduate students and postdocs, and some very fruitful collaborations.

The enzyme-design process began with quantum chemical calculations to work out what the ideal dimensions of the 'active site' — the part of an enzyme that brings the reactants together — would be for the chosen reaction. Baker's group lacked the expertise to do this, so he called upon Kendall Houk and his team at the University of California, Los Angeles.

Then, in Baker's own lab, graduate students Alexandre Zanghellini and Lin Jiang developed general algorithms for designing new enzymes, and postdoc Andrew Wollacott fine-tuned these algorithms to the reaction of interest. Next, postdocs Daniela Röthlisberger and Eric Althoff used the algorithms to construct computer models of potential enzymes, which were converted into DNA sequences. The team introduced the corresponding DNA fragments into the bacterium Escherichia coli to produce proteins, which were purified and their activities tested.

“Everyone had different backgrounds,” says Baker. “I am a strong believer in people with complementary backgrounds working together. Sometimes just in the act of articulating what you are doing you get better ideas.”

The next step was to carry out 'directed evolution' on those enzymes that looked promising. This involves randomly mutating enzymes and selecting those that show catalytic prowess. For the necessary expertise, Baker looked to Dan Tawfik, Olga Khersonsky and their co-workers at the Weizmann Institute of Science in Rehovot, Israel. “They've been doing directed evolution with naturally occurring enzymes for quite a while, and they're really good at it,” Baker says. “They took our designs and were able to find mutations that made the enzyme work much, much better.”

For Baker, teamwork doesn't just mean solving tough research challenges. “Hard problems are hard, so it takes a while to solve them, and there are a lot of downs before there are ups,” he says. “In a team, you share the lows and then you get to share the highs, and I think emotionally, that makes it easier to tackle problems.”

The results of the collaboration are a landmark achievement, but Baker isn't one to brag. “A lot of people put in a lot of hard work, but like most heads of labs, I didn't really do anything useful,” he says with a laugh. “I just walked around and talked to the people doing the work.”