Biosen-Based Growth-Coupled Evolution of Epimerases for Improved D-Allulose Biosynthesis

A Smarter Way to Evolve Epimerases

Enzyme engineers have a new trick. A Chinese research team built a biosensor that ties D-allulose production directly to cell growth. This turns evolution into a survival game. The best mutants win by making more D-allulose. No robotic picking. No expensive assays. Just growth.

The team applied this platform to improve ADAE, a ketose 3-epimerase that converts D-fructose into D-allulose. Their top mutant, M42, delivers a 6.28-fold boost in catalytic activity. Its half-life at 60°C stretches 2.5 times longer. And it converts 32.2% of D-fructose into D-allulose within three hours.

Background: Why D-Allulose Needs Better Enzymes

D-allulose is a rare sugar with nearly zero calories. It tastes like sucrose, but doesn't spike blood glucose. Food and beverage brands want it. The problem is supply. D-allulose exists in tiny amounts in nature. Industrial production usually relies on enzymatic isomerization of D-fructose. Ketose 3-epimerases do the job, but most native enzymes are too slow or too fragile for harsh process conditions.

Directed evolution can fix that. But screening mutants is tedious. Traditional screens measure activity in microtiter plates or use colorimetric assays. They take time and miss subtle improvements. The bottleneck isn't mutation. It's selection.

What the Study Did

The researchers designed a D-allulose-dependent biosensor called Biosen. This sensor recognizes D-allulose inside the cell and activates a growth-promoting reporter. When they mutagenized the epimerase gene, cells carrying better variants produced more D-allulose. More D-allulose meant stronger biosensor signal. Stronger signal meant faster growth.

That's growth-coupled evolution under selective pressure. The population of cells does the screening for you. Cells with poor enzymes stagnate. Cells with good enzymes dominate. Over successive rounds, the enzyme improves without manual intervention.

In this case, the workflow started with random mutagenesis of ADAE. The library was transformed into cells containing Biosen. Cells were grown on D-fructose. Only those with active epimerase variants could convert fructose to allulose, trigger the biosensor, and divide. After several rounds, the team isolated the M42 mutant.

Key Findings

The M42 mutation is not a small tweak. It's a meaningful step toward industrial viability.

Property	M42 mutant performance
Catalytic activity	6.28-fold increase vs. wild type
Thermostability at 60°C	Half-life extended 2.5-fold
D-Allulose yield	32.2% conversion in 3 hours

Thermostability matters most for manufacturing. Higher temperature operation speeds up reactions and reduces contamination risk. A 2.5-fold longer half-life at 60°C means more conversions per enzyme batch. That cuts catalyst costs.

The 32.2% conversion isn't the highest reported for KEases, but it's achieved in only three hours with a mutant selected entirely through growth coupling. The process can likely be optimized further with additional rounds or reaction engineering.

What This Means for Manufacturers

This paper changes the economics of enzyme discovery. You don't need a thousand-well plate robot to find better epimerases. A biosensor and a shake flask can do it. That lowers the barrier for smaller companies. It also accelerates development timelines for custom enzymes.

For ingredient buyers, the impact is indirect but real. Better enzymes mean higher D-allulose yields at lower cost. This can stabilize pricing as demand grows. It also opens the door to producing D-allulose from cheaper fructose streams.

The Biosen platform isn't limited to ADAE. The same principle can be adapted to other ketose 3-epimerases or related sugar-converting enzymes. Any reaction that produces a detectable metabolite could benefit from growth-coupled screening.

FAQ

What exactly is Biosen?
Biosen is a D-allulose-dependent biosensor. It detects intracellular D-allulose and triggers a growth reporter. Cells that produce more D-allulose grow faster.

Does this screening method work only for thermophilic enzymes?
No. The study used thermostability as a target, but the screen selects for any variant that increases D-allulose production. The 60°C half-life improvement was a bonus from the selection pressure applied during evolution.

How soon can this approach reach commercial enzyme development?
The research team already produced a significantly improved mutant in this study. The growth-coupled platform is straightforward to replicate in labs with standard molecular biology equipment. Commercial adoption is a matter of engineering scale-up, not conceptual feasibility.

The Next Step

The biosensor approach doesn't replace rational design or computational prediction. It complements them. You can create a focused library based on structural insights, then let growth-coupled selection find the winners. That's a faster path from gene to process.

The M42 mutant gives formulators a better starting point. But the bigger story is the method. Biosen proves that a simple growth-based assay can outperform elaborate screening infrastructure. For the D-allulose industry, that's a tool worth watching.