The Peptide in the Cheese: A Century of Nisin

Nisin became a food preservative because a dairy bacterium kept killing its neighbours. A 1928 report from dairy research described one strain of Streptococcus lactis inhibiting Lactobacillus bulgaricus, the organism used to make yoghurt, and the substance responsible was characterised, named and eventually manufactured 12. It is a peptide of 34 amino acids made by Lactococcus lactis, evaluated by the joint FAO and WHO expert committee on food additives in 1968, and listed in Europe as E234 45. This is the history of how that happened.

Nisin is also the cluster's reminder that a peptide antibiotic need not be a medicine. The other stories in this series are about drugs that went to hospitals. This one went to the dairy.

A nuisance in the yoghurt vat

Dairy fermentation is a managed collision of microbes. A starter culture of chosen bacteria is added to milk to sour it, flavour it and set it, and the cheese- or yoghurt-maker's skill lies in keeping the intended organisms in charge. A culture that fails is not mysterious but expensive. In the 1920s the dairy-research division of the United States Department of Agriculture was studying exactly this problem: what limited the growth of the organisms in a lactic fermentation 1.

The 1928 paper in the Journal of Bacteriology reported an observation of a kind that dairy bacteriologists knew well and rarely wrote up. A strain of Streptococcus lactis, a common milk-souring organism, inhibited Lactobacillus bulgaricus, one of the two organisms that make yoghurt 1. The finding was framed as a problem in culture management. A strain of one bacterium that suppresses another could ruin a batch, and the practical lesson was to avoid pairing them.

That is how most discoveries of antibiotic activity begin: as an inconvenience. The 1928 report did not describe a preservative, a medicine or even a molecule. It described an effect, and the effect would take twenty years to acquire a name.

It helps to remember what the field looked like. In 1928 nobody was looking for antibiotics in the modern sense; penicillin was a curiosity on a Petri dish in London that year, and the idea of a bacterial product as a tool was still more than a decade from respectability. The people who recorded the inhibition between two milk bacteria were not on the trail of a drug. They were trying to keep their fermentations working 1.

Naming it: 'group N inhibitory substance', 1947

Through the 1930s and 1940s, workers in Britain returned to the observation. The streptococci that soured milk were classified by serology into groups, and the ones that made the inhibitory agent belonged to group N. The agent was therefore called the group N inhibitory substance, and from that clumsy description the contraction nisin was derived 2.

The Lancet paper of 1947 by A. T. R. Mattick and A. Hirsch carries the name in its title, in brackets, which is the earliest place this piece was able to confirm it in print 2. It reports further observations on the substance and its activity, and sits in a period when the success of penicillin had made every research group in medicine alert to the possibility of a new antibacterial. The interest of the authors lay in whether the agent had a role in medicine.

Dorset: the first factory

It did not find that role. It found a role in food, and the route there ran through a small British company, Aplin and Barrett, which began producing nisin commercially in the 1950s and sold it under the name Nisaplin 3. The company is associated with Beaminster in Dorset, where a laboratory building bearing its name still stands, although this piece has not been able to confirm from the sources consulted exactly where the production took place.

What made a food use plausible was a problem that dairies actually had. Processed cheese, made by heating and blending natural cheese, can be spoiled by spore-forming bacteria that survive the heat and then grow in the finished product. A preservative that prevented their growth would reduce the loss, and nisin was good at exactly that. The review of its applications published in 1996 describes the use in processed cheese and in other foods 3.

The 1968 evaluation and an E-number

A substance cannot be added to the international food supply without being assessed. In 1968 the Joint FAO/WHO Expert Committee on Food Additives considered nisin at its twelfth meeting, and its evaluation, published in the series of technical reports numbered 430, records an acceptable daily intake for the antimicrobial preservative 5. The additive carries the international numbering system number 234, and in the European Union the E-number E234.

The European Food Safety Authority re-examined the additive in 2017, as part of its programme of re-evaluating old additives and in connection with a proposed extension of use 4. That is routine for an additive of this age. It is also a reminder that approval is not a final state: the dossier on a food ingredient is revisited as methods and expectations change.

Two things are worth noticing about this record. The first is how little of it concerns discovery: the committee was asked a question about consumption, not about origin, and the document says nothing of the dairy laboratory where the story began. The second is the unusual position nisin occupies in the regulatory landscape. A peptide that is an antibiotic by its biology is, by its legal classification, a food additive, with the obligations and the freedoms that go with that. It is not assessed as a drug, and it is not sold as one.

A lantibiotic: what makes the molecule unusual

Nisin is a polycyclic peptide of 34 amino acids. After the chain is built by the ribosome, enzymes in the producing bacterium modify it, converting some residues into unusual ones, including lanthionine, a sulphur-bridged amino acid, and dehydroalanine. The bridges close the chain into a series of rings. The class is named after lanthionine: lantibiotics, antibiotics that contain it 4.

Why bacteria have not resisted it as they resist drugs

A frequent claim about nisin is that bacteria have not developed the kind of widespread resistance to it that they have to clinical antibiotics, despite many decades of use. The claim needs care. Laboratory studies have produced nisin-resistant strains, and resistance mechanisms are known. What is generally said is that resistance has not spread through the food supply the way it has through hospitals 3.

Explanations offered in the literature include the way the molecule acts. Later work showed that nisin binds a building block of the bacterial cell wall, a target that is hard for a bacterium to change without cost, and it can also form pores in the membrane. Whether those are the sole reasons is a question for the microbiology literature rather than for this piece. A second factor is simply exposure: a preservative is used at low levels in foods that are eaten, not given in repeated courses to patients, and the selection pressure is of a different kind.

Nisin today: from processed cheese to the clinic that never quite happened

Nisin is used today in a variety of foods and, in many countries 3. Dairy remains the classic setting, with processed cheese, but the review also describes its use in other foods where spore-forming bacteria are a risk.

The medicine it might have been remains a footnote. Nisin has been studied in the laboratory as a possible therapeutic, and animal and in vitro work exists, but it did not follow the path of the drugs described elsewhere in this series into routine clinical use. Why it did not is a mixture of chemistry, since peptides of this kind are not well suited to injection, and economics. The history here belongs to food.

A useful way to hold both facts is to separate a molecule's biology from its category. The same chain of amino acids is an antibacterial agent to a microbiologist, a preservative to a cheese-maker and an additive number to a regulator, and none of those descriptions is wrong. Which one prevails depends on who is asking, and on what paperwork has been filed.

Where the food-science and microbiology literature takes over

The questions this article leaves open are technical ones. How much nisin a given food contains, how stable it is across processing and storage, which organisms are susceptible and how resistance arises are matters for food-science and microbiology papers, and for the regulatory documents cited here.

What belongs to history is the arc: an observation in a dairy culture in 1928, a name in 1947, a factory in the 1950s, a committee in 1968 and a number on a label. A peptide with no connection to medicine had become, quietly, one of the most widely eaten of all.