14 OCT 2026 — For more than 40 years chemists have argued about exactly how a single enzyme turns a simple chain of three amino acids into the strained, double-ring core of penicillin. An Oxford-led team has now recorded it with X-ray lasers, capturing atomic-resolution snapshots and intermediates the researchers themselves describe as unanticipated.
The study, published in Nature Catalysis on 9 October, is about basic chemistry rather than a new drug. Its authors argue it matters for antibiotics anyway, because understanding how nature builds these rings is the first step to persuading enzymes to build new ones.
The enzyme at the centre
The enzyme is isopenicillin N synthase, IPNS. Working with iron and oxygen, it takes the tripeptide ACV and closes two rings in one go: the four-membered beta-lactam ring that gives penicillins and cephalosporins their antibacterial power, and a second, sulphur-containing ring beside it. Stanford, whose X-ray laser did much of the work, calls IPNS essential for making all of the medically important penicillins.
Crystal structures of IPNS have existed since the 1990s, as the paper's references show, and the same Oxford group has used X-ray lasers on it before. But the structures of the intermediates that form the two rings, the paper says, "have remained elusive".
How you film a reaction
The team grew tiny crystals of IPNS already loaded with iron and ACV, but kept them from oxygen so the reaction could not start. Droplets of these crystals went onto a moving tape 2 millimetres wide, which carried them through an oxygen chamber and then into the path of an X-ray free-electron laser. Each pulse lasts only femtoseconds, short enough to capture a structure at room temperature.
Changing the tape speed changed how long each crystal had been reacting. At SLAC's Linac Coherent Light Source the team collected data at seven points between one and ten seconds, according to the methods section. A further ten-second point came from Korea's PAL-XFEL, and a structure of a stable stand-in for a fleeting state came from Japan's SACLA. Strung together, the snapshots form what Oxford's announcement calls a "molecular movie".
Two surprises along the way
One surprise is the thioaldehyde, the intermediate formed when the enzyme oxidises ACV's sulphur-bearing cysteine. The paper reports that it "unexpectedly does not coordinate the iron centre": the sulphur lets go of the iron at a point where many proposed mechanisms had it still attached. Experiments in which the team mutated the enzyme backed that up.
The second came from feeding the enzyme a version of ACV carrying deuterium, a heavier form of hydrogen, and that made it possible to observe a monocyclic beta-lactam: the first ring closed, the second not yet formed, and the sulphur bonded to the iron again. The authors conclude that an iron(IV) species, helped by water molecules in and around the active site, drives the final ring closure. Movements in parts of the protein far from the active site also appear to steer the chemistry.
Why it matters for antibiotics
"Penicillin has shaped modern medicine, but there is still much to learn about how nature builds this important antibiotic structure," said Christopher Schofield, a professor of chemistry at Oxford and a senior author. First author Patrick Rabe said that understanding the process in atomic detail would let researchers "begin to think about engineering these enzymes to produce new or improved antibiotic scaffolds".
A method with wider reach
That is a long road from a crystal structure, and the paper itself makes no claim of a new medicine. Its more immediate reach may be wider than penicillin: IPNS belongs to a large family of iron-dependent enzymes that includes ones human cells use to sense oxygen, and Oxford says the approach can be used to study fleeting steps in other enzymes. The study also adds to a quiet lineage at Oxford, where Dorothy Hodgkin solved the structure of penicillin by X-ray crystallography in 1945.
View author profile → · Editorial policy
