Mary Kable, Department of Food Science and Technology, University of California Davis, USA
Mary E. Kable1, Yanin Srisengfa, Miles Laird, Jose Zaragoza, Jeremy McLeod2, Jessie Heidenreich2 and
Maria L. Marco1
1. Department of Food Science and Technology, University of California Davis;
2. Hilmar Cheese Company, Hilmar, CA
Background: Raw milk harbors diverse bacteria that can have significant impacts on the quality and safety of fluid milk and dairy products. Currently, the bacterial composition in raw milk after transport in tanker trucks and within commercial processing facilities is not well understood. Therefore, we set out to identify the bacterial composition of raw milks entering and moving through large-scale, dairy product-manufacturing plants in California.
Methods: Raw bovine milks were collected from tanker trucks arriving at two dairy processors in the California San Joaquin Valley during three seasons (spring, summer and fall) and from five large volume silos during two days in the summer season. Additionally, a total of 142 milks were collected in duplicate once every 1.5 to 4.5 hours during spring (71) and late summer (71) from 9 and 11 types of equipment, respectively, within the early milk processing steps including storage, pasteurization, separation and ultrafiltration. Half of each of the milks collected from within the processing facilities were treated with propidium monoazide (PMA) to prevent DNA from dead bacterial cells from being PCR amplified and sequenced. Illumina MiSeq sequencing of the 16S rRNA V4 region was performed to characterize the bacterial diversity in these milks.
Results: Examination of milk from 899 tankers revealed highly diverse bacterial populations. However, even with this complexity, a core microbiome was present consisting of 29 taxonomic groups and high proportions of Streptococcus, Staphylococcus and unidentified members of Clostridiales. Comparisons between the raw tanker milks also showed seasonal differences. Raw milk collected in the spring contained the most diverse bacterial communities with the highest total cell numbers but were also the most homogenous, with the lowest variation between the different tankers tested. Transfer of the milk to storage silos resulted in distinct shifts in the bacteria present, with some of the silos containing significantly higher proportions of Acinetobacter than the tankers used to fill them. Bacterial communities within the facility were also susceptible to seasonal variation that mirrored changes in the raw tanker milks. In addition to natural variability, the equipment used for short-term storage tended to have the greatest impact on determining the bacterial community structure. The feed tanks before pasteurization, concentrate silos after ultra filtration and standardized silos formed the most distinct clusters out of all equipment compared by PCoA of weighted UniFrac distances. PMA treatment of pasteurized samples resulted in a trend in reduced Streptococcus, Staphylococcus and Corynebacterium suggesting that the majority of these organisms do not survive pasteurization, although damaged cells still linger in the product. Pseudomonas was among the bacterial taxa that tended to be present in greater relative abundance after pasteurization, suggesting post-pasteurization contamination.
Conclusion: The bacterial community in raw milk is highly variable although a core set of organisms is always present. This variability in combination with the tendency for standing milk to develop outgrowths of particular taxa that are not completely controlled by pasteurization presents a challenge for dairy producers in controlling the quality of their product. This work helps to illustrate the exact nature of that challenge and these results can be applied to modify processing methods to improve the consistency of high quality dairy products.Download PDF