Food microorganisms & starter cultures

A starter culture is a known population of organisms added deliberately at a known dose. Its purpose is reproducibility: the same acidification curve, the same flavour compounds, the same texture, batch after batch. Everything interesting about starter cultures follows from what you give up when you replace an ecosystem with a defined population.

Defined and undefined

An undefined mixed culture — a traditional whey or backslopped culture — contains many strains whose identities are not fully catalogued. It is robust, because if one strain is knocked out the others carry the fermentation, and it is inconsistent for the same reason. A defined culture is a specified blend of characterised strains. It is reproducible, controllable and specifiable, and it is fragile in one particular way.

What the culture is actually doing

Acidification is the primary function, and its rate matters as much as its endpoint. In cheesemaking, acid development has to stay in step with rennet coagulation and curd handling, because pH at each stage governs how much calcium leaves the casein micelle, which in turn governs the texture of the finished cheese. A culture that acidifies too fast or too slowly makes a different cheese, not merely a faster or slower one.

Beyond acid, cultures do secondary work: proteolysis, releasing peptides and free amino acids that become flavour precursors; lipolysis; citrate metabolism producing diacetyl, the buttery note of cultured butter and some cheeses; and CO₂ production, which is what makes eyes in certain cheeses. Texture in fermented dairy also depends on exopolysaccharides secreted by some strains, which thicken the product without added hydrocolloids.

Phage is the defining vulnerability

Dairy fermentation is run at enormous volumes on a small number of well-adapted strains, in plants that cannot be sterilised, with a fresh substrate arriving daily. That is close to ideal conditions for bacteriophage. A phage attack on the dominant strain stops acidification, and a stalled vat is both a lost batch and a safety concern.

Defence is structural rather than chemical: rotating unrelated strains so no single phage has a persistent host, selecting strains with different phage-receptor profiles, hygienic plant design, and screening cultures against phage isolated from the factory itself.

Lactic acid bacteria have their own defences, and one of them changed biology generally. CRISPR arrays in Streptococcus thermophilus were shown to be an adaptive immune system that acquires spacers from the phage that attacked it — work that came out of dairy culture research and became the foundation of genome editing.

Getting a live organism into a sachet

Cultures are supplied frozen or freeze-dried, and both are lethal if done carelessly. Freezing damages membranes through ice crystal formation and osmotic stress; drying removes the water that maintains membrane and protein structure. Cryoprotectants and lyoprotectants — sugars such as trehalose and sucrose, and polyols — substitute for water at the membrane surface and vitrify, limiting the damage. Survival is strain-specific and finite, which is why cultures carry a viable count and an expiry rather than just a weight.