Cells replicate by dividing in two, but with each replication, small variations creep into the DNA sequence. Rather than waiting for natural mutations, Arnold causes small mistakes to be made when DNA is copied in the test tube. Then she inserts all the mutated copies into living microbes, which translate the genes into proteins. At that point, she combs through the different proteins to find the ones she likes.
Arnold opted for this brute-force approach, running hundreds—even thousands—of experiments with random mutations in the proteins, selecting those with the characteristics she wanted to breed in the second generation and so on over multiple generations. And because microbes reproduce every 20 minutes, it didn’t take billions of years to see the results. “All you have to do is look at antibiotic resistance to understand how quickly biology can adapt,” she says.
She published her first papers on evolutionary protein engineering in the early 1990s in the face of considerable skepticism. A bit of intellectual snobbery may have been in the mix. There is an unspoken hierarchy in academia. It’s a culture that favors curiosity-driven basic research over practical applications, and it draws a fine distinction between science and engineering. “Some people looked down their noses at it,” Arnold admits. “They might say ‘It’s not science’ or that ‘Gentlemen don’t do random muto-genesis.’ But I’m not a scientist, and I’m not a gentleman, so it didn’t bother me at all. I laughed all the way to the bank, because it works.”
More than one senior colleague over the years has accused her of being arrogant, a charge she cops to readily and does not apologize for. It’s a quality that has helped her stay on top in a highly competitive research environment. “I’ve always been confident,” she says, describing herself as “honestly arrogant—I wasn’t pretending. I never had any doubt about the utility and importance of what I was doing.”
Among the many intriguing results from her research were new proteins that don’t break down under high temperatures, making them ideal for everything from laundry detergents to developing new drugs. Merck drew on her techniques to develop a process for manufacturing the diabetes drug Januvia.
She also developed new kinds of enzymes that help convert cellulose in plants into sugars and then into useful fuels and chemicals. That work, and a passion for building a green chemicals industry based on renewable resources, led to Arnold co-founding in 2005 a biofuels company called Gevo, which went public a few years later. But she didn’t become the CEO—the job she’d envisioned for herself as an undergraduate. By then she knew it wasn’t her forte. “I know how to do science. I know how to make things,” she says. “I don’t know how to run a company. Now that’s a really tough job.”
Arnold’s scientific gamble back in the 1980s to jump into biotechnology and engineer the biological world has paid off in hundreds of publications, dozens of patents, and countless professional accolades. She is one of only a few people with membership in all three of America’s national academies: the National Academy of Sciences, the Institute of Medicine, and the National Academy of Engineering. In January, she went to the White House to receive the National Medal of Technology and Innovation.
Many academic researchers take a sabbatical break every seven years or so, but Arnold has taken just one in her 27 years at Caltech. She spent that time traveling with her family through Australia, Madagascar, Namibia, South Africa, and Egypt.
She has faced plenty of challenges. Right after she returned from her sabbatical, she was diagnosed with breast cancer and spent two years battling it into remission. She lost her husband, Caltech physicist Andrew Lang, in 2010. Today she is cancer-free, focused on her work, and getting her three sons through high school. “I don’t sit around feeling sorry for myself,” she says. “There’s always somebody who’s a lot worse off than you. I love what I do, and I’m grateful for every day I can do it.”
Arnold may have a pragmatic bent, but ultimately she’s just as eager as her colleagues to decipher nature’s code when it comes to how genetic sequence dictates function—she’s simply realistic about the unlikelihood of this happening in her lifetime. “We can do all this manipulation, but we don’t understand the rules of composition for DNA,” she says. “So the best we can do is cut and paste pieces from compositions that nature has already written.” If she can’t compose, she’ll go right on breeding. Maybe she hasn’t been doing this for 4 billion years, but Arnold just might give nature a run for its money yet.