PS6-14. Comparison of Sampling Methods for Characterization of the Equine Uterine Microbiome During Postpartum Involution.

Abstract Maintaining a 12-month foaling interval in mares requires conception soon after parturition, given the approximately 340-d length of gestation. However, uterine involution takes approximately 28-d, creating overlap in the process of uterine recovery and establishment of the new pregnancy. The uterine microbiome may influence pregnancy success during this period; however, the microbiome during uterine involution has not been characterized and optimal sampling techniques remain unclear. Therefore, the objective of this study was to compare three sampling techniques for characterizing the uterine microbiome during uterine involution hypothesizing that a sterile, double-guarded brush would yield better sequencing results. To test this, 8 gestating, multiparous Quarter Horse mares (14±3 yr) were maintained on pasture and fed to meet or exceed nutrient requirements during gestation. Upon signs of parturition, mares were moved into individual foaling stalls (3.66 m × 7.32 m) and monitored 24 h until parturition. Samples were collected at 12 and 72 h postpartum. Prior to sampling, mares' vulvas were cleaned with a soft soap a minimum of 3 times. Then, a double-guarded sterile swab was introduced into the vagina as a positive control to account for vaginal contamination in sampling. Then, a double-guarded sterile swab, a double-guarded sterile brush, and a nonguarded infusion pipette for low-volume lavage (LVL; 150 mL Lactated Ringer's solution), were introduced independently, in that order, through the vagina and cervix into the uterus. All samples were collected using a sterile gloved hand with the sampling apparatus guarded by a second sterile glove when introduced to and exiting the uterus. Once collected, samples were placed in a sterile conical vial, snap frozen in liquid nitrogen, and stored at -80 °C. DNA was extracted using the DNeasy PowerSoil Pro Kit and sequenced by Novogene for the V3-V4 hypervariable region of the 16S rRNA encoding gene. Raw sequencing reads were processed in QIIME2 and DADA2 using NovaSeq standard operating procedures. Alpha diversity (Shannon; P > 0.73) and beta diversity (Weighted Unifrac; P > 0.35) did not differ among sampling methods at either time point, after post-hoc comparisons using the Tukey's HSD test. Firmicutes (³42.6%) and Proteobacteria (³22.8%) were the most abundant phyla across all sample methods, which aligns with previous findings in the healthy, non-pregnant mare. Low-volume lavage frequently failed to recover the full infused volume because of the enlarged postpartum uterus and was therefore not considered optimal for future studies. In conclusion, the uterine microbiome can be detected in the postpartum involuting uterus, and characterization was consistent across sampling methods. The similarity between sampling techniques suggests that a double-guarded sterile swab or brush can be used for downstream DNA analysis; however, LVL had suboptimal application.