Biosensor-Driven Growth-Coupled Evolution Improves Epimerase Performance for D-Allulose Biosynthesis

If you spend time developing enzymatic routes to rare sugars, you know the usual grind. You mutate an enzyme, pick hundreds of colonies, run assays, and pray for one decent hit. This study flips that process. Instead of screening each variant individually, the researchers made the cells do the selection themselves.

D-allulose is a low-calorie sugar with real market pull. The bottleneck is the enzyme. Ketose 3-epimerases, or KEases, convert D-fructose into D-allulose, but natural versions are often too slow or too fragile at industrial temperatures. The authors engineered those enzymes with a clever growth-coupled biosensor system and ended up with a mutant that is clearly better.

The screening problem

Enzyme engineering for D-allulose has a practical issue. You can create huge libraries of mutants, but finding the few improved variants is like looking for a specific grain of sand on a beach. Standard microtiter plate screening takes time, labor, and expensive reagents. The team behind this paper wanted to compress that process into something faster and more direct.

Their answer was a D-allulose-dependent biosensor. The biosensor is designed to detect D-allulose inside the cell and convert that signal into a growth advantage. A cell that produces more D-allulose from D-fructose gets a stronger signal. A stronger signal makes that cell grow faster. Over successive rounds, the better mutants outgrow the weaker ones. No one has to pick and assay every single clone.

What the study did

The researchers built a growth-coupled in vivo screening platform for KEases. The core principle is simple: link D-allulose production to survival. When a mutant epimerase works, the cell produces D-allulose, the biosensor registers it, and growth is favored. When the enzyme is weak, growth stalls. Over time, the population shifts toward the most productive variants.

They applied this platform to directed evolution of ADAE, a ketose 3-epimerase. The goal was to push both catalytic activity and thermostability, because both matter in a real production process. Activity drives conversion rates. Thermostability determines how long the enzyme can keep working at elevated temperatures.

Key findings

The evolution campaign paid off. The best variant, M42, showed large improvements over the starting enzyme. To put it plainly: this is the kind of result that makes a process engineer sit up.

Performance marker	Wild-type ADAE baseline	M42 mutant
Catalytic activity	Reference	6.28-fold higher
Half-life at 60 °C	Reference	2.5-fold longer
D-allulose conversion from D-fructose in 3 hours	Reference	32.2%

The 32.2% conversion number matters. It shows that the enzyme can produce allulose from fructose in a meaningful yield within a few hours. The 60 °C half-life extension matters just as much. Higher temperature tolerance means less enzyme required, fewer contamination risks, and longer operating windows.

What this means for manufacturers

For ingredient buyers and formulators, the relevance is not academic. Enzymes are a major cost driver in rare sugar production. A variant with six times the catalytic activity and 2.5 times the thermostability at a common processing temperature changes the cost picture. Reactor residence times can drop. Enzyme loading can drop. Or the same equipment can push through more batches per shift.

This screening platform also points to faster catalyst development cycles. If you can run evolution continuously on a growth-based readout, you can cycle through more generations in less time. For allulose suppliers, that may mean more consistent supply, better enzyme options, and lower production costs down the road.

There is also a practical takeaway for formulators evaluating allulose ingredients. Enzyme evolution informs price volatility. A breakthrough like M42 may not appear on an ingredient label, but it can appear in contract pricing, availability, and supplier reliability. When producers get a hotter, tougher enzyme, the savings eventually move through the supply chain.

FAQ

What exactly is a growth-coupled biosensor screen?
It is a selection system where the product you want, in this case D-allulose, creates a growth advantage. Cells producing more allulose grow faster, so the best enzyme mutants naturally win the race. The researcher does not need to assay every culture.

Does M42 work at industrial scale?
The paper reports laboratory evidence with strong performance metrics. The 3-hour conversion and improved 60 °C half-life are promising, but scale-up depends on feedstock quality, reactor design, enzyme immobilization, and downstream recovery. Still, the starting point is far better than the wild type.

Is this platform specific to this one enzyme?
The biosensor is specific to D-allulose, and the evolution method targets KEases. That makes it a natural fit for allulose production. The same growth-coupled logic could be adapted to other sugar epimerases if a suitable biosensor exists for their product.

M42 is not the final enzyme, but it shows the direction. Selection pressure did the scoring, and the result is a catalyst with better kinetics and better staying power. For anyone waiting on cheaper or more reliable allulose supply, that is exactly the sort of progress worth watching.