Researchers design protein scissors that cut specific targets

Overview

Updated 54 minutes ago

University of Washington researchers have designed protein-cutting enzymes from scratch. The synthetic proteases snip a chosen protein at a chosen point, and they work inside living human cells.

Two papers in Nature describe the work. A cysteine-based design cleaved its target up to 30 million times faster than the uncatalyzed reaction; a zinc-based design was 100 million times faster. If the approach matures, it could destroy misfolded proteins behind Alzheimer's disease and ALS, or neutralize bacterial toxins that no drug currently targets.

Why it matters

These programmable protein scissors could destroy disease proteins — Alzheimer's plaques, ALS aggregates, bacterial toxins — that no existing drug can target.

Questions about this story

Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.

No questions yet — be the first to ask.

Key Indicators

19%

Cysteine protease design success rate

13 of 69 designed enzymes cut their target sequence.

36%

Zinc protease design success rate

More than one-third of designed enzymes showed activity in a single design round.

100 million×

Fastest bond-cleavage acceleration

Zinc protease sped peptide bond hydrolysis more than 100 million times over the uncatalyzed reaction.

Nature papers published

Covering cysteine and zinc metalloprotease designs.

Voices

Curated perspectives — historical figures and your fellow readers.

Ever wondered what historical figures would say about today's headlines?

People Involved

David Baker

Director, UW Medicine Institute for Protein Design; Professor of Biochemistry, University of Washington School of Medicine

Led the protease design effort; shared the 2024 Nobel Prize in Chemistry

Hojae Choi

Doctoral student in biochemistry, University of Washington School of Medicine

Lead author of the cysteine protease study

Anqi Chen

Lead author of the metalloprotease study

Designed zinc proteases that cut amyloid-beta

Kejia Wu

Co-lead author of the metalloprotease study

Co-designed zinc proteases using the RoseTTAFold Diffusion model

Organizations Involved

UW Medicine Institute for Protein Design

Research institute

Conducted the protease design research

University of Washington institute that computationally designs proteins, including the new synthetic proteases.

University of Washington School of Medicine

Academic medical school

Home of the researchers and the Institute for Protein Design

Academic medical school whose department of biochemistry houses the protein design work.

Timeline

3 events Latest: 2 days ago

October 2026
UW Medicine announces designed molecular scissors
Latest Announcement

The institute reports the proteases publicly, noting they cut targeted proteins precisely and work inside living human cells, including against proteins tied to Alzheimer's and ALS.
2 days ago
Nature publishes cysteine protease design paper
Publication

Researchers report de novo designed enzymes using an activated cysteine to cut peptide bonds. 13 of 69 designs worked, up to 30 million times faster than the background reaction.
October 2026
Nature publishes zinc metalloprotease design paper
Publication

Designs using a bound zinc ion as the cutting blade showed activity in 36% of cases. The most active cleaved amyloid-beta 100 million times faster than the uncatalyzed reaction.
October 2026

Scenarios

Designed proteases reach human trials

Possible Resolves by End of 2029

Discussed by: UW Medicine researchers; the Nature papers note therapeutic promise

Researchers continue testing the proteases on disease targets, first in animal models. If safety and efficacy hold, a designed protease aimed at a neurodegenerative disease protein — amyloid-beta in Alzheimer's or TDP-43 in ALS — advances to a Phase 1 clinical trial.

Designed proteases become a standard research tool

Likely Resolves by Q2 2028

Discussed by: The Nature papers and UW Medicine announcement describe the work as a step toward tailor-made molecular scissors

Independent labs adopt the designed proteases for controlled protein cleavage in experiments — protein purification, studying protein degradation, and building synthetic circuits. Wide use in basic research follows, much as early CRISPR tools spread through labs.

Clinical development stalls on delivery and immune hurdles

Possible Resolves by End of 2029

Discussed by: Ars Technica and GeekWire coverage note delivery, immunogenicity, and switchability as barriers

Proteins are hard to deliver inside cells, may trigger immune responses, and the proteases need on/off control. If these obstacles persist, the enzymes stay research reagents and no therapy advances.

Historical Context

3 moments from history that rhyme with this story — and how they unfolded.

June 2012 – 2013

CRISPR-Cas9 genome editing (2012–2013)

In 2012, researchers including Jennifer Doudna, Emmanuelle Charpentier, and later Feng Zhang showed that a bacterial immune system called CRISPR-Cas9 could be repurposed to cut specific DNA sequences at will. A guide RNA directs the Cas9 enzyme to a precise spot in the genome.

Then

Within months, laboratories worldwide adopted CRISPR for gene editing, and it became the standard method for altering DNA.

Now

CRISPR underlies a growing field of gene therapies and engineered crops, though delivery and off-target effects remain hurdles.

Why this matters now

Like designed proteases, CRISPR is a programmable molecular scissors. It shows how a lab tool that cuts biological molecules can grow into a therapy platform — and the delivery and specificity problems it had to overcome.

October 2024

2024 Nobel Prize in Chemistry for protein design

The Royal Swedish Academy awarded the 2024 Nobel Prize in Chemistry to David Baker for computational protein design, and to Demis Hassabis and John Jumper for protein structure prediction. Baker's lab had built software that designs new proteins folding into desired shapes.

Then

The prize recognized protein design as a mature field with practical uses in medicine and materials.

Now

It cemented algorithmic protein design as a foundation for creating new molecular tools, including enzymes.

Why this matters now

This protease work extends the capability that won Baker the prize — from designing static protein shapes to designing proteins that perform chemistry.

2008 – 2010s

First de novo designed enzymes (2008–2010s)

In 2008 the Baker lab reported the first de novo designed enzyme, a Kemp eliminase that accelerates a simple reaction with a low energy barrier. Later designs tackled similar model reactions.

Then

De novo enzymes were built, but they were limited to activating relatively easy chemical bonds.

Now

The new Nature papers note that de novo enzyme design had largely succeeded only on low-energy-barrier model reactions — until now.

Why this matters now

Cleaving a peptide bond, which is stable in water for years, is a far harder target than the model reactions earlier de novo designs achieved.