Investigating the cow skin and teat canal microbiomes of the bovine udder using different sampling and sequencing approaches

Abstract

There is a need for standardized, efficient, and practical sampling methods to support large population-based studies of the internal and external epithelial microbiomes of the bovine udder. The primary objective of this study was to evaluate different sampling devices for the isolation of microbial DNA originating from the internal and external teat epithelium. Secondary objectives were to survey and compare the microbial diversity of external and teat canal epithelial microbiomes using amplicon and shotgun metagenomic sequencing approaches. To address these objectives, we enrolled a convenience sample of 24 Holstein dairy cows and collected samples from the external epithelium at the base of udder, the external teat barrel epithelium, the external teat apex epithelium, and the teat canal epithelium. Extracted DNA was quantified and subjected to PCR amplification of the V4 hypervariable region of the 16S rRNA gene and sequenced on the Illumina MiSeq platform (Illumina Inc., San Diego, CA). A subset of samples was subjected to a shallow shotgun metagenomic assay on the Illumina HiSeq platform. For samples collected from the external teat epithelium, we found that gauze squares consistently yielded more DNA than swabs, and Simpson's reciprocal index of diversity was higher for gauze than for swabs. The teat canal epithelial samples exhibited significantly lower diversity than the external sampling locations, but there were no significant differences in diversity between teat apex, teat barrel, and base of the udder samples. There were, however, differences in the microbial distribution and abundances of specific bacteria across external epithelial surfaces. The proportion of shotgun sequence reads classified as Bos taurus was highly variable between sampling locations, ranging from 0.33% in teat apex samples to 99.91% in teat canal samples. These results indicate that gauze squares should be considered for studying the microbiome of the external epithelium of the bovine udder, particularly if DNA yield must be maximized. Further, the relative proportion of host to non-host DNA present in samples collected from the internal and external teat epithelium should be considered when designing studies that utilize shotgun metagenomic sequencing.

Characterization of rumen, fecal, and milk microbiota in lactating dairy cows

“…Trueperella pyogenes is a common summer mastitis pathogen ( Ishiyama et al, 2017 ), and was present in a single sample at high abundance, indicating a potential subclinical IMI. Ornithinimicrobium has been detected previously in milk and bedding microbiome samples ( Dean et al, 2021 ; Li et al, 2021 ), and was linked to be significantly higher in sand bedding in comparison to other beddings ( Ray et al, 2022 ), which is the type of bedding used in the present study.…”

The composition and natural variation of the skin microbiota in healthy Australian cattle

“…Our study is the rst to investigate this using 25 herds of Australian cattle. The most abundant bacterial phyla (Proteobacteria, Firmicutes, Actinobacteria and Bacteroidetes) which were observed across all herds (Figure S1), have been previously reported in healthy bovine digits and teat skin [1,19,20]. Moraxellaceae, Sphingomonadaceae, Bacillaceae, Burkholderiaceae and Peptostreptococcaceae are bacterial families generally present on the teat and digit skin of healthy cattle [18, [21][22][23][24].…”

The microbiome of common bedding materials before and after use on commercial dairy farms

“…However, such low biomass samples require careful consideration in microbiome studies given the possibility of contamination from extraction kit reagents, especially PCR master mix and even molecular biology grade water [ 40 – 42 ], which can sometimes exceed the abundance and diversity of the resident microbiome [ 41 ]. To control for this, we included negative controls and used them to identify and remove likely contaminants from the sequence data [ 43 , 44 ]. Despite these internal controls, it is important to note that cross-contamination could still explain some of the extreme variability in microbial composition of the data obtained from these samples, particularly within unused NSA samples (Figs.…”