Researchers at the UW Medicine Institute for Protein Design have built "molecular scissors" from scratch - synthetic proteases that cut other proteins at precise, chosen points. Published in Nature (1, 2), the achievement marks a step from predicting protein shapes toward designing machinery that physically changes matter in targeted, programmable ways.
Most medicines work through binding: a small molecule or antibody attaches to its target and stays there, blocking or activating it one molecule at a time. This one–to–one relationship means treating a million copies of a disease target requires roughly a million drug molecules. That is why many protein medicines must be given in large quantities by infusion.
Enzymes work on a different scale. An enzyme changes its target and releases, unchanged and ready to act again. A single designed protease can cut thousands to millions of specified targets over its lifetime, allowing the same effect at a substantially lower dose.
The studies report designs that selectively cleave human disease–associated proteins, without destroying the protein's essential structured functions. This opens approaches impossible with binding agents: destroying viral or bacterial proteins outright, clearing misfolded proteins that accumulate in neurodegenerative disease, or snipping protective caps off drugs so they activate only in needed tissues.
Two families of designed proteases
One study created cysteine proteases built around an activated cysteine amino acid as the cutting blade. These achieved a 19 percent success rate and cut target sequences up to 30 million times faster than the uncatalyzed reaction. The resulting proteins adopt folds unlike any naturally occurring protease.
The second study designed metalloproteases that use a bound zinc metal ion rather than an amino acid as the blade. Thirty–six percent of these designs showed activity.
Designing proteases from scratch had remained elusive because the blades need near–atomic precision to enable cutting chemistry, and natural proteases rely on subtle flexibility to grip their targets. These advances came from newer computational models and deep learning methods that improved designers' ability to navigate such complexities. Artificial intelligence (AI) -powered tools generated and refined candidate structures, then filtered them through predictive models to identify designs likely to fold and function as intended.
"This is a step toward designing tailor–made molecular scissors," said a researcher. "Proteases are just the beginning," added another. "We envision a new generation of designed molecules that can act on biological systems in increasingly sophisticated ways."