# Scientists turn blank protein barrels into working enzymes

> A study in Nature Chemical Biology reports that researchers at the University of Bayreuth and the University of Ottawa have transformed inactive artificial protein scaffolds into working enzymes. The team used a computational workflow called CANVAS to add a custom catalytic pocket...

Canonical URL: https://www.argo.net/scientists-turn-blank-protein-barrels-into-working-enzymes/
Byline: University of Bayreuth
Published: 2026-07-07T04:50:10+00:00
Categories: Chemistry, News

![Vibrant closeup of a colorful molecular model illustrating abstract scientific concepts](https://www.argo.net/wp-content/uploads/2026/06/enzyme_molecular_model.jpg)

A study in **Nature Chemical Biology** reports that researchers at the [University of Bayreuth](https://www.nature.com/articles/s41589-026-02250-w) and the University of Ottawa have transformed inactive artificial protein scaffolds into working enzymes. The team used a computational workflow called CANVAS to add a custom catalytic pocket to minimal protein barrels, creating engineered enzymes that could carry out a benchmark chemical reaction with unusually high activity.

The result tackles one of protein design's stubborn goals: making enzymes from the ground up. Enzymes drive chemistry in living cells with speed and precision. If scientists can design them reliably, they could build cleaner ways to make medicines, materials and industrial chemicals.

The research was led by Prof. Dr. Birte Höcker, chair of Biochemistry III at the University of Bayreuth, in collaboration with Prof. Dr. Roberto Chica's group at the **University of Ottawa**. Their work focuses on a famous protein shape known as the TIM barrel, a structure that nature has used again and again to power enzyme reactions.

## A new route to artificial enzymes

Artificial proteins can now be designed with remarkable structural accuracy. Scientists can ask a computer to generate a protein fold, then test whether the real molecule adopts that shape in the lab. That progress has changed protein engineering, yet enzyme design has remained much harder.

An enzyme needs more than a stable shape. It needs a small working region where the chemical reaction happens. This region must place atoms in the right orientation, hold the reacting molecule and stabilize fleeting chemical states during the reaction.

The new study shows that minimal artificial protein scaffolds can be upgraded with such a working region. The researchers used **de novo proteins**, meaning proteins designed from scratch rather than copied from natural enzyme sequences.

Dr. Julian Beck, lead author of the study and a researcher in Höcker's group, said the team combined several computational methods. "This enabled us to specifically extend the artificial TIM barrels to include a tailor-made active site."

## Why TIM barrels matter

The **TIM barrel fold** is one of biology's most successful enzyme architectures. It appears in roughly 10% of known enzymes and can support many kinds of chemical reactions. Its name comes from triosephosphate isomerase, one of the classic enzymes where this fold was studied.

In simple terms, a TIM barrel resembles a molecular cylinder built from repeating protein elements. Natural versions often include loops and lids that help form a pocket for chemistry. Those add-on features make the barrel useful as an enzyme, since they help bind a target molecule and position catalytic residues.

Researchers had already designed artificial TIM barrels on computers. Experiments confirmed that these proteins could fold into the desired shape. Their limitation was function, since these minimal structures lacked the detailed active regions found in natural enzymes.

That gap made TIM barrels a strong test case. If a blank artificial barrel could be turned into an enzyme, the same idea might help researchers build many other tailor-made catalysts from simple protein frameworks.

## How CANVAS builds an active site

The team's workflow is called **CANVAS**, short for customizing amino acid networks for virtual active-site scaffolding. Its purpose is to take a minimal scaffold and add the molecular features needed for catalysis.

The study abstract describes the core strategy this way: "Here, we present CANVAS, a computational workflow that introduces a structural lid into a minimal de novo TIM barrel." That lid helps anchor catalytic residues and shape the pocket where a reaction can occur.

First, the researchers placed a model of the desired chemical reaction onto the TIM barrel scaffold. Then they designed new protein segments that could form a lid over the catalytic face. Finally, they refined the surrounding amino acids so the pocket could hold and stabilize the reacting molecule.

This approach matters because the **active site** is where an enzyme's precision emerges. A stable protein fold provides the framework. The carefully arranged pocket provides the chemistry.

The team also checked whether the designs matched reality. According to the Nature Chemical Biology study, crystal structures of selected variants helped confirm that designed features appeared in the expected places.

## The test reaction that proved it worked

The researchers tested their designs with the **Kemp elimination**, a well-known benchmark reaction in artificial enzyme design. It is useful because it can be measured clearly in the laboratory and has long served as a comparison point for new enzyme-design methods.

Prof. Dr. Birte Höcker explained the choice directly: "We chose the Kemp elimination as a test reaction for the new enzymes." The reaction gave the team a practical way to compare their new proteins with earlier engineered catalysts.

Starting from two scaffolds, the researchers designed nine variants with tailored lids for the Kemp elimination. Four showed measurable activity. That outcome showed that the added lids and redesigned pockets could do more than preserve structure; they could support catalysis.

The reaction itself is synthetic in this context. Its value comes from being a clean test of whether a designed protein can accelerate a specific chemical transformation. For protein designers, that kind of benchmark helps separate promising architecture from decorative molecular geometry.

## What KempTIM1 and KempTIM4b showed

The strongest first-round result was **KempTIM1**. In the Nature Chemical Biology study, KempTIM1 reached a catalytic efficiency of 21,000 M−1 s−1. The researchers reported that this was seven times higher than comparable first-round Kemp eliminases in recent publications.

That performance came without further experimental optimization. In practical terms, the first computational design round already produced an enzyme-like protein with strong measurable activity. For a field that often needs repeated rounds of mutation and screening, that is a notable step.

The team then used structural information from a lower-activity variant to guide further design. This ensemble-based optimization produced a much stronger version. The paper reports an increase of more than 1,600-fold for that design path, reaching 32,000 M−1 s−1.

The public summary identifies the improved variant as **KempTIM4b**. Its activity exceeded that of KempTIM1, showing that the CANVAS-designed scaffolds could be refined after the first build.

Together, these results suggest that minimal TIM barrels can serve as starting platforms for efficient designed enzymes. The work also shows the value of pairing computational design with structural biology, since measured structures can reveal how to improve a working molecule.

## Why this could matter for green chemistry

Enzymes are attractive tools for **green chemistry** because they often work in water, at moderate temperatures and with high selectivity. Those traits can reduce waste and energy use in chemical manufacturing.

Designed enzymes could expand that advantage beyond the reactions nature already performs. Researchers want catalysts tailored for industrial chemistry, biotechnology, medicine, environmental cleanup and materials production. Reliable de novo enzyme design would give scientists a broader toolkit.

Beck framed the potential in those terms, saying, "This opens up new possibilities in biotechnology and green chemistry." The immediate study remains a proof of principle for a benchmark reaction. Its broader importance lies in the workflow it demonstrates.

The CANVAS strategy also offers a modular idea. Instead of searching nature for a protein that almost fits a desired job, researchers can start with a well-behaved scaffold and build the missing catalytic features into it. The Nature Chemical Biology results show that this can produce active enzymes from previously inactive artificial barrels.

Future work will need to test the method across more reactions and more demanding chemical tasks. For now, the study gives enzyme designers a clearer path from a blank protein shape to a working catalyst, one custom-built pocket at a time.