Growth-Coupled Evolution Pressure with Biosen Improves Epimerase for D-Allulose Biosynthesis

A practical answer to an enzyme bottleneck

D-allulose formulators know the problem. The enzyme that converts D-fructose into D-allulose is often too slow. It also falls apart under heat. A new paper describes a clever fix: a growth-coupled screening platform built around a D-allulose-dependent biosensor called Biosen. The team used it to evolve a ketose 3-epimerase (KEase). The winning mutant, M42, shows 6.28-fold higher catalytic activity. Its half-life at 60°C extends 2.5-fold. It converts D-fructose to D-allulose at 32.2% yield in three hours. Those numbers matter.

Why KEases hold back allulose production

D-allulose is a rare sugar with a sucrose-like taste and very few calories. Producing it at scale usually means enzymatic isomerization of D-fructose. KEases catalyze that step. But natural KEases rarely meet industrial demands. High temperatures help speed reactions and reduce contamination. Heat destroys many enzymes. So processors face a tradeoff: push temperatures and lose activity, or keep temperatures low and wait longer.

Better enzymes solve that tradeoff. The challenge is finding them. Directed evolution can generate thousands of variants. Screening them one by one is slow and expensive. This study takes a different route.

What the study did

The researchers built an in vivo screening platform. The system uses Biosen, a biosensor that responds to D-allulose. The sensor links D-allulose production to cell growth. Cells that produce more D-allulose grow better. That creates a growth-coupled selection pressure. No manual picking of individual colonies. No complicated assays at every step. The best enzyme variants enrich themselves through growth.

The team applied this platform to KEases. They generated mutant libraries and let the biosensor sort them. The selection pressure targeted catalysis directly. That is the key point. Many screening methods catch improved protein stability or binding, but miss activity. This one rewards actual D-allulose formation.

Key findings

The evolved mutant M42 outperformed the starting enzyme across the board.

Metric	Result
Catalytic activity	6.28-fold higher than parent
Half-life at 60°C	2.5-fold longer
D-allulose from D-fructose	32.2% conversion in 3 hours

These are not marginal gains. A 6.28-fold activity jump means less enzyme needed per batch. A 2.5-fold longer half-life means more usable production time at elevated temperature. And 32.2% conversion in three hours gives process designers a real starting point for yield optimization.

What it means for manufacturers

This work is about the catalyst, not the finished sweetener. But catalysts decide cost. If engineered KEases like M42 reach commercial scale, D-allulose production becomes more efficient. Lower enzyme loads cut material costs. Better thermostability widens process windows. Higher conversion reduces downstream purification load. Each of those shifts improves the economics of D-allulose.

For ingredient buyers, this signals progress. Enzyme suppliers are moving toward smarter screening tools. Growth-coupled biosensors shorten development cycles. They also make it possible to evolve enzymes for specific process conditions, not just generic lab conditions. That is exactly what manufacturers need.

The Biosen platform may also extend beyond this one mutant. The design targets KEases broadly. Any enzyme in that family could benefit from the same growth-coupled selection. That opens the door to more thermostable and more active variants for different production setups.

FAQ

What is Biosen? Biosen is a D-allulose-dependent biosensor. It couples D-allulose concentration to cell growth. Cells that produce more D-allulose grow faster, so selection happens automatically.

Is M42 ready for commercial food production? Not yet. The study demonstrates a promising engineered enzyme. Scale-up, process validation, and regulatory approval still lie ahead.

Can this approach improve other rare sugar enzymes? The platform targets ketose 3-epimerases, the enzyme family that makes D-allulose. The same biosensor-based selection could likely be adapted to related KEases.

The takeaway

Enzyme evolution used to be slow. You mutated, screened, and hoped. This work replaces hope with a growth-linked selection system. For allulose makers, that means better catalysts are on the way. And better catalysts mean a more affordable supply of D-allulose for your products.