Scientists at the University of Washington have achieved a significant breakthrough in enzyme design, led by recent Nobel Prize winner David Baker. The research team has developed artificial intelligence tools capable of creating custom enzymes from the ground up, marking a major advance in addressing what they describe as one of science’s most formidable challenges.

The groundbreaking research, published in Science, demonstrates a novel approach to engineering these crucial proteins that catalyze chemical reactions in living organisms. The team’s work represents a fundamental shift from traditional enzyme design methods, which typically involved combining existing protein components in hopes of achieving desired functions.

“This new approach allows us to create enzymes specifically designed for any reaction we’re interested in,” explains Anna Lauko, who recently completed her Ph.D. in Baker’s laboratory and served as one of the study’s co-lead authors alongside Sam Pellock and Kiera Sumida.

The researchers focused their efforts on developing a serine hydrolase, an enzyme type known for its ability to break specific chemical bonds found in various carbon-containing molecules, including plastics and human fats. They utilized RFdiffusion, an open-source AI model previously developed by Baker’s lab, in conjunction with a new tool called PLACER to identify the most promising enzyme candidates.

Pellock, an acting instructor in the lab, compared previous enzyme design methods to searching for a suit at a thrift store – unlikely to find a perfect fit. The new approach, however, allows for the creation of custom-designed proteins that more precisely match their intended molecular targets.

While the artificially designed enzymes haven’t yet matched the performance of their naturally occurring counterparts, they represent some of the most effective synthetic enzymes created to date. The achievement suggests scientists are approaching the capability to engineer enzymes that could potentially surpass nature’s versions for specific human applications.

One promising application being explored by team member Sumida involves developing enzymes to address plastic pollution. Given that synthetic plastics are relatively new materials in evolutionary terms, natural enzymes haven’t had time to evolve effective mechanisms for breaking them down. The team’s technology could lead to
custom-designed enzymes capable of degrading various types of plastic waste.

Baker, who received the 2023 Nobel Prize in Chemistry for his pioneering work in protein design and AI applications in molecular biology, directs the Institute for Protein Design at UW Medicine. The research team included contributors from multiple UW departments, including biochemistry, biological physics, computer science, and molecular engineering, as well as collaboration with UCLA’s Department of Chemistry and Biochemistry.

The breakthrough could have far-reaching implications for various industries, from pharmaceutical manufacturing to environmental remediation. With this new capability to design custom enzymes for specific reactions, researchers anticipate an increase in successful enzyme design projects in the coming years.

This scientific achievement builds upon decades of previous attempts at enzyme design that often yielded disappointing results. The new AI-driven approach opens up possibilities for creating precise, effective enzymes that could address various technological and environmental challenges facing society.

The research represents a convergence of artificial intelligence and molecular biology, demonstrating how computational tools can accelerate scientific discovery and innovation in ways that were previously impossible. As these technologies continue to evolve, they may enable the development of increasingly sophisticated and effective custom-designed enzymes for a wide range of applications.