A soil bacterium makes lasalocid A, an antibiotic used widely in agriculture. To build this molecule, the microbe must perform a feat of chemical choreography: it takes a floppy chain-like precursor and folds it into a rigid cage structure by installing oxygen atoms at exactly the right positions, in exactly the right three-dimensional orientations.

Get it wrong, and the molecule fails.

Now researchers have captured high-resolution snapshots of the enzyme responsible, revealing how a single protein manages to perform two consecutive oxygen-addition reactions on the same substrate while maintaining flawless control over the shape of the final product. The findings illuminate a decades-old mystery in natural product biosynthesis and could accelerate efforts to design new antibiotics through biological engineering.

Lasalocid A belongs to a family called polyether polyketides. These molecules are defined by multiple ring-shaped ether groups embedded in their carbon scaffolds. They work by wrapping around metal ions like potassium or sodium and smuggling them across cell membranes, disrupting the delicate ion balance that cells depend on to survive. This makes polyethers potent antimicrobials.

Farmers use lasalocid A and related compounds to prevent coccidial infections in poultry and livestock. But toxicity issues have kept polyethers out of human medicine. Modifying their structures could solve this problem, yet chemical synthesis is prohibitively difficult because polyethers contain numerous stereocenters—positions where atoms can be arranged in mirror-image configurations, only one of which produces the desired biological effect.

Understanding how bacteria build polyethers naturally could enable scientists to engineer microbial factories that produce safer, more effective variants.

The enzyme at the heart of this study is Lsd18, a flavin-dependent monooxygenase from Streptomyces lasalocidi. During lasalocid A assembly, Lsd18 receives a long, flexible polyketide chain from upstream biosynthetic machinery. This chain contains two carbon-carbon double bonds at specific locations. Lsd18's job: convert both double bonds into epoxides—three-membered oxygen-containing rings—with the R,R stereochemical configuration at each site.

Critically, Lsd18 must act on two chemically distinct positions within the same molecule and ensure that both reactions yield the same three-dimensional geometry. Six of lasalocid A's ten chiral centers trace back to these epoxidation events.

How does one enzyme accomplish this?

The research team solved three X-ray crystal structures of Lsd18: the enzyme alone with its cofactor FAD, the enzyme bound to a substrate mimic, and the enzyme bound to a product mimic. Together, these structures provide a molecular movie of the catalytic process.

Lsd18 contains an unusually large substrate-binding pocket—approximately 35 angstroms deep and 23 angstroms wide at its broadest point. This spaciousness is key. Unlike enzymes that grip their substrates in a single fixed orientation, Lsd18 allows its polyketide substrate to adopt different shapes inside the pocket. The flexibility permits the enzyme to position first one double bond, then the other, next to the reactive flavin cofactor.

But size alone doesn't explain stereoselectivity.

Adjacent to the active site sits a smaller subpocket lined by the amino acids tyrosine-218, valine-252, and valine-342, with isoleucine-72 nearby. In the substrate-bound structure, this subpocket cradles an ethyl group attached to the carbon-carbon double bond undergoing epoxidation. This interaction is not merely passive. It dictates which face of the double bond points toward the flavin.

The geometry is unforgiving. When the ethyl substituent nestles into the subpocket, the re face of the double bond is exposed to the reactive oxygen species generated by the flavin cofactor. Oxygen transfer from this angle produces an R,R epoxide. The alternative orientation—where the si face would be exposed—is sterically forbidden because the ethyl group would collide with isoleucine-72.

Both double bonds in the substrate carry ethyl substituents at equivalent positions. Both therefore bind in the same orientation. Both yield R,R epoxides.

To test this model, researchers created mutant versions of Lsd18 in which key subpocket residues were altered. They then measured the stereochemical outcome of epoxidation reactions using a substrate analogue that cyclizes spontaneously after oxygen addition, allowing the researchers to infer which epoxide geometry had been formed.

Wild-type Lsd18 produced only the R,R product. Every mutant—I72A, Y218F, V252A, and V342A—produced mixtures of R,R and S,S epoxides. Disrupting the subpocket abolished stereocontrol. The mutants also showed reduced overall activity, indicating that these residues not only guide stereochemistry but also promote catalysis.

The structures also clarified how Lsd18 performs sequential epoxidations. Molecular modeling revealed that the enzyme pocket is large enough to accommodate the substrate in two distinct conformations: a J-shape in which the terminal double bond approaches the flavin, and a V-shape in which the internal double bond does so. In both poses, the ethyl substituent fits into the subpocket, ensuring consistent stereochemistry.

Whether the substrate is released and rebinds between the two epoxidation events, or remains bound throughout, is unknown. The order in which the two double bonds are modified is also unclear. These questions remain open.

The stereocontrol mechanism appears to be conserved across polyether biosynthesis. Sequence alignments of monooxygenases from lasalocid A, monensin, nanchangmycin, and salinomycin pathways show that the subpocket residues are highly conserved. Yet different enzymes produce different stereochemical outcomes.

Lsd18 and its counterpart from salinomycin biosynthesis generate exclusively R,R epoxides, while the monensin and nanchangmycin enzymes produce both R,R and S,S epoxides. The difference correlates with substrate structure. Trisubstituted double bonds—those bearing a methyl or ethyl branch—become R,R epoxides because the branch engages the subpocket. Disubstituted double bonds lack such a branch and can bind in either orientation, yielding mixed products.

The findings complete a structural trilogy. Over the past decade, the same research group has determined atomic structures for three core enzymes in lasalocid A biosynthesis: the polyketide synthase that assembles the carbon backbone, the epoxide hydrolase that converts epoxides into cyclic ethers, and now the monooxygenase that installs the epoxides. Together, they validate a 40-year-old theoretical model for how bacteria build polyether antibiotics.

That model, proposed in 1983, posited that all ionophore polyethers arise through a common biochemical logic: build a linear polyketide, add oxygen to form epoxides with precise stereochemistry, then trigger a cascade of ring-closing reactions to form the ether rings. The structural data now confirm this stepwise choreography at atomic resolution.

For protein engineers, the results offer both promise and caution. Lsd18's broad substrate pocket suggests it could accept a variety of polyene substrates, raising the possibility of feeding it non-natural precursors to generate novel polyether structures. But the enzyme's dependence on specific interactions with its partner polyketide synthase complicates matters. Mixing and matching components from different biosynthetic pathways may require redesigning protein-protein interfaces, not just active sites.

Stability is another obstacle. Lsd18 tends to precipitate during extended reactions. The researchers addressed this by engineering a double mutant, T189M/S195M, that expresses at higher levels, shows improved thermal stability, and resists digestion by the protease trypsin—all while retaining full catalytic activity. Further rounds of directed evolution could yield even more robust variants suitable for industrial applications.

The stakes extend beyond antibiotics. Polyether scaffolds appear in numerous bioactive natural products, and the ability to modify them systematically could unlock new drugs for cancer, parasitic diseases, and microbial infections. Chemical synthesis remains slow and expensive. Biological synthesis, guided by structural blueprints like those reported here, offers a faster route.

It also offers precision. Lsd18 doesn't just add oxygen. It adds oxygen in a specific three-dimensional arrangement, using a subpocket as a molecular jig to align the substrate with atomic exactness. That kind of control is difficult to achieve through traditional chemistry.

The polyether biosynthesis machinery represents one of nature's more intricate molecular assembly lines. Now, piece by piece, its inner workings are coming into focus.

Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1002/anie.202504982