Science1 publisher2 min readPublished
Simulations link the ligands a Bacillus enzyme holds to how open its shape is
Shibaura Institute of Technology simulations of Bacillus subtilis PyNP retained bound ligands in 75.6% of closed-shape trajectories against 55.1% of open ones. The result informs nucleoside-analog drug design, though it measures how ligands bind and stops short of the reaction itself.
The Scientist · Science desk

What happened
- The team ran all-atom molecular dynamics across four conformational states, using 13 structural probes and trajectories totaling 15.6 microseconds, and published the work in ACS Omega.
- Between its most closed and most open states, PyNP's active-site pocket expanded by roughly 1.4-fold. That sets the size of the space a ligand has to occupy.
- The enzyme's sugar preference among ribose- and 2'-deoxyribose-containing compounds varied; which sugar it favored depended on the specific compound and the conformation it was in.
- The residue Tyr165 acted as a moving lid over the active site, and deleting it in extra simulations cut ligand retention. That result supports the lid mechanism at the computational level.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- capability Shape-dependent capture gives groups designing inhibitors and enzymatic synthesis routes a design handle, and the authors tie it to greener manufacturing.
- constraint For making nucleoside analogs the useful readout is the chemical step, and these runs stop at binding; connecting the two is still ahead.
- precedent The work offers a framework for treating enzymes as multi-shape targets in recognition studies, inhibitor design and enzyme engineering.
The puzzle came first. For years, groups using PyNP to make artificial nucleosides have wondered why swapping the sugar changes the reaction [15]. "We wanted to understand this question from the perspective of atomic-scale motions that experiments alone cannot reveal," Hatano said [11].
The model measures holding, not turnover. It counts how long a ligand stays bound over a trajectory; whether the enzyme converts it to product is a separate question. The closed-to-open retention gap works out to about 20 percentage points [1]. That number does not show how the holding maps onto the chemical step a chemist wants when building a nucleoside analog, and the group points to further calculations and experiments to make the link [13].
The sugars sorted by ring shape. In the simulations ribose sat in the North (C3'-endo) pucker for 64.3% of the sampled time, while 2'-deoxyribose sat South (C2'-endo) for 57.9% [7]. The split was widest for the unsubstituted compound and shrank as the substituent at the 6-position grew larger [8].
The lid result comes with a limit the authors state themselves. The strongest Tyr165 signal came from a closed structure of a related species, not B. subtilis, so a closed-state B. subtilis structure is still needed to confirm it [10].
"Enzymes should not be viewed as single, static structures when we think about molecular recognition," Hatano said. "Our results show that different conformations can change how compounds are retained and how sugar-related preferences emerge" [12].
PyNP is used to make anticancer, antiviral and antibacterial candidates and artificial nucleosides [1], and to make fluorescently labeled and stable-isotope-labeled nucleic acids [14]. Understanding how its shape governs capture is what the authors offer toward inhibitor design and cleaner enzymatic synthesis [14].
What to watch
- Whether the group's promised further calculations and experiments connect the measured binding behavior to PyNP's actual catalytic steps.
- Whether a closed-state Bacillus subtilis structure confirms the Tyr165 lid; the signal now rests on a related-species structure.
- Whether the conformation-dependent sugar preference shows up as measurable yield differences in wet-lab nucleoside synthesis.