Uterine microbial profiles in healthy postpartum dairy cows do not vary with sampling techniques or phases of estrous cycle

In this study, we aimed to compare uterine microbial profiles in postpartum dairy cows, determined by bacteriological culture and next-generation sequencing, using three uterine sampling techniques (swab, cytobrush, and lavage) and induced phases of the estrous cycle (estrus and diestrus). Fifteen healthy postpartum dairy cows at 53 +/- 5 days postpartum were enrolled in the study. Uterine samples were collected during a fixed-time artificial insemination protocol. Viable bacteria were aerobically cultured from part of each sample, and bacterial isolates were identified through Sanger sequencing of the 16S rRNA gene. Total genomic DNA was extracted from the remainder of undiluted samples to quantify bacterial load using 16S rRNA qPCR and characterize the microbiome by metagenomic sequencing of the V1-V3 region of the 16S rRNA gene. Microbial profiles and composition were analyzed using the Shannon-Weaver diversity index and principal component analysis, respectively. Out of 87 samples, 88 % (77/87) were culture positive. The proportion of culture-positive uterine samples did not differ between sampling techniques (P = 0.39) or estrous cycle phases (P = 0.99). However, swab, cytobrush, and lavage techniques yielded 1.5, 9 and 9 times greater bacterial loads (P < 0.01), respectively, during diestrus than estrus phase. Moreover, during diestrus phase, the cytobrush method yielded 3 and 6 times more bacteria (P < 0.01) than both the lavage and swab methods. The most abundant bacterial genera identified from both bacteriological culture and metagenomic sequencing were Bacillus and Enterococcus, regardless of sampling technique or phases of the estrous cycle. Bacterial genera in moderate to low abundance through metagenomic sequencing included Streptococcus, Oscillospiraceae, and Lachnospiraceae. Notably, the uterine microbial profiles and composition, determined by metagenomic sequencing, did not differ by sampling techniques (P = 0.55 and P = 0.60, respectively) or estrous cycle phases (P = 0.34 and P = 0.17, respectively). In conclusion, our results suggest that any of the sampling techniques can be reliably used to study the uterine microbiome of healthy cows at random phases of the estrous cycle. However, it is important to consider potential differences in bacterial yield as a confounding factor.