Tracking the dairy microbiota from farm bulk tank milk to skimmed milk powder using 16S rRNA gene amplicon sequenicng and shotgun metagenomic sequencing

Microorganisms from the environment can enter the dairy food chain at multiple stages between milk collection and the end of processing, with potential associated quality and safety-related implications. The ability to track these microorganisms can be greatly enhanced by the use of high throughput DNA sequencing (HTS). Here HTS approaches, both 16S rRNA gene amplicon and shotgun metagenomic sequencing, were applied to investigate the microbiome of fresh, early and late lactation (i.e., seasonal), milk samples collected from farm bulk tanks, collection tankers, whole milk silos, skimmed milk silos, a cream silo, and powder samples to investigate the microbial changes throughout dairy processing. 16S rRNA gene analysis established that the microbiota of raw milks from farms and in collection tankers was very diverse but that the psychrotrophic spoilage-associated genera, Pseudomonas and Acinetobacter, are present in all samples. Upon storage within the whole milk silo at the processing facility, these genera, corresponding to Pseudomonas fluorescens and Acinetobacter baumannii species as determined by shotgun metagenomic sequencing, become dominant. The milk powder generated during the early lactation period had a microbial composition that was very different from that of the starting raw milk in that two thermophilic genera, Thermus and Geobacillus, selected for within the processing plant, were dominant. In contrast, the microbiota of powder generated during the late lactation period more closely resembled raw milk inputs, and was dominated by spoilage-associated psychrotrophic bacteria. This study demonstrates that HTS can be used to trace microorganisms from raw milk through processing to final powdered products, classify them to species level and, in the process, highlights that different sources of contaminating microbes can be relevant.