Microbiota Assessments for the Identification and Confirmation of Slit Defect-Causing Bacteria in Milk and Cheddar Cheese

Validated methods are needed to detect spoilage microbes present in low numbers in foods and ingredients prior to defect onset. We applied propidium monoazide combined with 16S rRNA gene sequencing, qPCR, isolate identification, and pilot-scale cheese making to identify the microorganisms that cause slit defects in industrially produced Cheddar cheese. To investigate milk as the source of spoilage microbes, bacterial composition in milk was measured immediately before and after high-temperature, short-time (HTST) pasteurization over 10-h periods on 10 days and in the resulting cheese blocks. Besides HTST pasteurization-induced changes to milk microbiota composition, a significant increase in numbers of viable bacteria was observed over the 10-h run times of the pasteurizer, including 68-fold-higher numbers of the genus Thermus. However, Thermus was not associated with slit development. Milk used to make cheese which developed slits instead contained a lower number of total bacteria, higher alpha diversity, and higher proportions of Lactobacillus, Bacillus, Brevibacillus, and Clostridium. Only Lactobacillus proportions were significantly increased during cheese aging, and Limosilactobacillus (Lactobacillus) fermentum, in particular, was enriched in slit-containing cheeses and the pre- and post-HTST-pasteurization milk used to make them. Pilot-scale cheeses developed slits when inoculated with strains of L. fermentum, other heterofermentative lactic acid bacteria, or uncultured bacterial consortia from slit-associated pasteurized milk, thereby confirming that low-abundance taxa in milk can negatively affect cheese quality. The likelihood that certain microorganisms in milk cause slit defects can be predicted based on comparisons of the bacteria present in the milk used for cheese manufacture. IMPORTANCE Food production involves numerous control points for microorganisms to ensure quality and safety. These control points (e.g., pasteurization) are difficult to develop for fermented foods wherein some microbial contaminants are also expected to provide positive contributions to the final product and spoilage microbes may constitute only a small proportion of all microorganisms present. We showed that microbial composition assessments with 16S rRNA marker gene DNA sequencing are sufficiently robust to detect very-low-abundance bacterial taxa responsible for a major but sporadic Cheddar cheese spoilage defect. Bacterial composition in the (pasteurized) milk and cheese was associated with slit defect development. The application of Koch’s postulates showed that individual bacterial isolates as well as uncultured bacterial consortia were sufficient to cause slits, even when present in very low numbers. This approach may be useful for detection and control of low-abundance spoilage microorganisms present in other foods.