UW's Institute for Protein Design uses AI to build molecular scissors aimed at Alzheimer's and ALS proteins

Researchers at UW Medicine's Institute for Protein Design in Seattle have used AI models developed at the institute to design proteases, enzymes that cut specific proteins at specific places, aimed at targets associated with Alzheimer's disease and ALS. The work was published in two papers in Nature this week.

ADVERTISEMENT

ARQS OFFERIs your business AI-ready?Get a focused AI Readiness Audit with practical next steps.$99 one-timeStart the audit →

Unlike many drugs that bind one target and stay there, the proteases can cut target after target. "There's this potential to use your drug at very, very low concentrations," said postdoctoral fellow Anqi Chen, a first author of one paper.

One team designed proteases that use a metal ion as the cutting blade; more than a third precisely cut an Alzheimer's-linked protein, 100 million times faster than without the enzymes. A second team, led by graduate student Hojae Choi, used the amino acid cysteine as the blade, with a 19% success rate and reactions up to 30 million times faster.

"There's super exciting applications," said institute director David Baker, the 2024 Nobel laureate in chemistry and senior author of both papers. The researchers are now working on targets for malaria, chronic disease and cancer, though much work remains, including keeping the immune system from attacking the enzymes, before they could become medicines.