How do bacteria close the rings of antibiotics? A new idea

Tetracyclines are antibiotics made by soil bacteria. The bacteria first build a long, flexible chain of carbon atoms with many oxygen atoms along it. Chemists call such a chain a polyketide.

An enzyme then folds the chain and closes it into rings, one after another. The enzyme for the first rings is called an aromatase/cyclase. Without it, the same chain still forms rings, but in the wrong places, and you get a mix of useless products.

Making tetracyclines in the lab is much harder. The best known synthetic route needs many steps and very cold reactions. A route that copies the bacteria could be shorter and greener.

Every ring closure in the chain is an aldol reaction, a classic way to join two carbon atoms. It needs two proton moves. First a proton leaves a carbon atom, which makes that carbon reactive. After the new bond forms, a proton must come back to an oxygen atom.

A proton is the nucleus of a hydrogen atom. It is tiny and moves fast, but where it comes from and where it goes decides how fast and how cleanly the reaction runs.

Crystal structures of two of these enzymes, TcmN and WhiE, show the folded chain inside a deep pocket. The researchers who solved them proposed that one water molecule, held in place by the protein, takes the first proton and later gives a proton back (Ames et al. 2008, Lee et al. 2012).

Not every enzyme works this way. A related enzyme, SnoaL, moves the protons with a single amino acid and no water (Sultana et al. 2004).

The preprint proposes that, outside the enzyme, the protons do not have to travel in one jump. They can hop along a chain of water or alcohol molecules held together by hydrogen bonds. Each molecule passes a proton to the next, like a bucket brigade. Chemists call this proton hopping, or the Grotthuss mechanism, after the scientist who first described it for water in 1806.

In this picture the single water in TcmN is the shortest possible chain. Longer chains could form from solvent molecules or from oxygen groups on the folded chain itself. The same relay could also hand a proton from one ring closure to the next and so link the rings into a cascade.

The preprint shows the steps on a simple model molecule and compares the idea with the usual textbook picture, in which a base grabs the proton directly.

The preprint lists four tests. Swapping normal hydrogen for heavy hydrogen (deuterium) should slow the reaction in a typical way, and mixtures of water and heavy water should reveal how many protons move at once. Changing the solvent should change the speed even when the base stays the same. Computer models should show a lower energy barrier for the relay than for a direct jump. And deuterium placed at one ring should show up at later rings if protons are passed along.

A computer study is planned in stages. It starts with small chains of water around a simple molecule and then moves to three real targets, including a polyketide used by the chemist Karsten Krohn.

If the idea holds, chemists could control ring closures by tuning the solvent around the reaction as well as the base. Flow reactors, where liquids are mixed in thin tubes under tight control, are well suited to that. This could help build new tetracycline-type antibiotics with fewer steps and less waste.

The work is part of sail.black research in chemistry, next to studies such as can fluorine form triple bonds and the CARATE molecule model.

This is a hypothesis paper. It contains no new lab results and no new computer results. The drawings use a textbook model molecule, not a real antibiotic precursor.

Some details are still open, among them the exact structures and solvents for the planned computer study. A full literature database search is also still needed before the idea can be called new. The preprint has not been peer reviewed.