Frances Arnold

Frances Arnold: Letting Evolution Do the Chemistry

At eighteen she was driving a Yellow Cab through Pittsburgh's steep hills, having already worked as a lunch-counter waitress, a pizza hand, a department store clerk and a cocktail waitress while living alone in a third-floor apartment. Four decades later she was in Stockholm collecting the Nobel Prize in Chemistry for an idea her colleagues had once dismissed as intellectual laziness: that if you cannot design an enzyme, you should breed one.

A Pittsburgh Education, Mostly Outside the Classroom

Frances Hamilton Arnold was born on 25 July 1956 in East Pittsburgh, Pennsylvania, the only daughter among five children of William Howard Arnold, an experimental physicist who took his doctorate at Princeton in 1955 and worked on nuclear reactor technology at Westinghouse, and Josephine Inman Routheau. Her grandfather, a general of the same name, had commanded American forces in Austria after the Second World War. Her father, often away, was her intellectual model: she thought him "the smartest person in the world because he knew all the answers, could explain how everything worked."

She wanted to be a transplant surgeon, inspired by Christiaan Barnard, until she discovered that blood made her queasy. By thirteen she had lost patience with classrooms altogether. She was sent briefly to a girls' school in Baltimore, then attended an inner-city public high school, and at Allderdice High in Squirrel Hill she hitchhiked to antiwar protests in Washington and supported herself through a rotating cast of jobs. It was an education in independence more than in chemistry, which she had not yet studied.

Princeton, Brazil, and a Red Volkswagen

She entered Princeton in 1974, the year the first women were graduating, into a class about fifteen per cent female and a mechanical and aerospace engineering department with far fewer. She kept driving cabs, worked in the library and cleaned house for the philosopher Thomas Kuhn. The oil shocks of the decade, and mentors who insisted science should answer to society, turned her toward alternative energy. She graduated magna cum laude in 1979.

What followed reads like deliberate refusal of a career track. She travelled from Ecuador to São Paulo on roughly two dollars a day for a solar energy internship with José Goldemberg, later Brazil's environment minister, spending six weeks crossing the Inca trail by bus, including a thirty-six-hour leg from Lima to Ayacucho sharing space with a flea-covered goat. Her first real job, from 1979, was at the Solar Energy Research Institute in Golden, Colorado, developing passive solar heating and cooling; she paid her rent by looking after horses.

When Ronald Reagan's election dimmed the prospects for solar research, she loaded a 1971 red Volkswagen Beetle and drove to Berkeley in January 1981 for a doctorate in chemical engineering. Her advisor Harvey Blanch steered her from cellulosic biofuels, whose funding had evaporated, toward the protein separation problems then confronting the new biotechnology industry. She studied affinity chromatography and the miseries of keeping proteins intact through manufacture, and learned biochemistry, enzymology and immunology from Jack Kirsch, Judith Klinman and Allan Wilson, at one point auditing organic chemistry by working as the official note-taker. She finished in 1985.

The Heresy of Not Designing

Arnold arrived at Caltech in mid-1986 as a visiting associate, learned site-directed mutagenesis in Jack Richards's lab, made the first mutant cytochrome c proteins for Harry Gray's electron-transfer work, and became an assistant professor of chemical engineering in January 1987.

The orthodoxy of the day was rational design: understand a protein's structure well enough, and you could reason your way to the mutations that would give it the property you wanted. Arnold watched it fail repeatedly, particularly on the problem of making enzymes work in organic solvents. So she inverted the method. Using error-prone PCR to scatter random mutations through a gene, she generated large libraries of variants in bacteria, screened them fast for the behaviour she wanted, kept the winners, and did it again. Evolution, run at laboratory speed, with the chemist choosing the selection pressure.

Protein chemists objected on two grounds: that most mutations are neutral, and that the beneficial ones she found sat nowhere near the active sites where the chemistry happens. Some called the whole approach lazy. Arnold's reply cut to the logic. "The argument seemed to be that because Nature never did it, it could not be done," she wrote. "But in fact, that was precisely why it could be done, and why it might even be easy to do so." By 1990 it was working.

What Directed Evolution Built

She took problems from Procter & Gamble, Degussa and Dow to prove the method on industrial reality rather than model systems. In 1996 she met Pim Stemmer, who had arrived at the same idea independently, and together they founded Maxygen. In 2005 she co-founded Gevo with Matt Peters and Peter Meinhold to engineer organisms that make liquid fuels; it now produces renewable jet fuel from biomass using engineered yeast. Her enzymes went into laundry detergents in place of harsher chemicals, into agricultural and paper manufacturing, into the production of type 2 diabetes drugs, and in 2009 into breaking down plant cell walls to turn agricultural waste into fuel.

Elected to the National Academy of Engineering in 2000 at forty-three, the same age at which her father had been elected, she and William Arnold are believed to be the only father-daughter pair in it. She later became the first woman elected to all three branches of the National Academies.

The Riskier Decade

A year-long family sabbatical through Australia, South Africa, Egypt, Namibia, Madagascar and Britain in 2003-04 was, she said, "the best year of my life." She returned at the end of 2004 to a diagnosis of breast cancer that had reached her lymph nodes, and worked daily through two surgeries and eighteen months of chemotherapy and radiation. In January 2010 her second husband, the cosmologist Andrew Lange, took his own life, leaving her three sons aged seventeen, thirteen and eleven. Her father died in 2015; her middle son, William, died in 2016 at twenty.

Her response in the laboratory was to take bigger risks. In 2012 her group built the first carbene transferase and nitrene transferase enzymes, catalysing reactions that do not exist anywhere in biology. Her bacteria have since been made to forge silicon-carbon bonds, boron-carbon bonds and bicyclobutanes: chemistry life never invented.

Why Frances Is Called a Genius

Arnold's distinction is not calculating power or mathematical depth. It is a specific and rarer thing: knowing when to stop trying to understand a system and start exploiting the fact that you don't. Every instinct of a well-trained protein chemist in 1990 said that engineering meant design, and design meant understanding. Arnold saw that the search space was too large and too counterintuitive for human reasoning, and that a mechanism already existed for navigating exactly such spaces. Her insight was epistemological before it was chemical, and it took nerve, because for several years it made her look like someone who had given up on thinking. Her own formulation is characteristically deflationary: "Life, the biological world, is the greatest chemist, and evolution is her design process. I may not be the best chemist but I do appreciate evolution."

The Nobel Committee honoured her in 2018 for "the directed evolution of enzymes," and she took half the prize, sharing the rest with George Smith and Gregory Winter for related work on phage display. Caltech's provost David Tirrell called it "a beautiful example of an enterprise that has both deep scientific significance and enormous practical consequences"; her colleague Jacqueline Barton said she had "broadened the repertoire of nature's catalysts."

The honest counter-case is worth stating. Smith and Winter reached comparable evolutionary methods independently, and Stemmer arrived at directed enzyme evolution in parallel, so the idea was in the air rather than uniquely hers. Directed evolution is also, by construction, a method that succeeds without generating deep mechanistic understanding: it produces enzymes, not explanations. Arnold would not dispute either point. What she claims, and what the record supports, is that she was first to make it work reliably, and stubborn enough to keep going when the field called it a shortcut.

Legacy

Arnold is the Linus Pauling Professor of Chemical Engineering, Bioengineering and Biochemistry at Caltech, where she was only the ninth woman ever hired to the faculty and is its first female Nobel laureate. She was the first woman to win the Charles Stark Draper Prize and the Millennium Technology Prize, and received the National Medal of Technology and Innovation from President Obama. Nearly sixty family members and former students came to Stockholm. Her method is now standard practice in laboratories that no longer remember it was ever controversial, and the enzymes it produces are quietly displacing toxic industrial chemistry, which is roughly what the nineteen-year-old riding a bus across the Andes to study solar power had in mind.