Plaza Dávila, María

Lactic acid bacteria (LAB) dominate human vaginal microbiota and inhibit pathogen proliferation. In other mammals, LAB do not dominate vaginal microbiota, however shifts of dominant microorganisms occur during ovarian cycle. The study objectives were to characterize equine vaginal microbiota in mares by culture-dependent and independent methods and to describe its variation in estrus and diestrus. Vaginal swabs from 8 healthy adult Arabian mares were obtained in estrus and diestrus. For culture-dependent processing, bacteria were isolated on Columbia blood agar (BA) and Man Rogosa Sharpe (MRS) agar. LAB comprised only 2% of total bacterial isolates and were not related to ovarian phases. For culture-independent processing, V3/V4 variable regions of the 16S ribosomal RNA gene were amplified and sequenced using Illumina Miseq. The diversity and composition of the vaginal microbiota did not change during the estrous cycle. Core equine vaginal microbiome consisted of Firmicutes, Bacteroidetes, Proteobacteria and Actinobacteria at the phylum level. At the genus level it was defined by Porphyromonas, Campylobacter, Arcanobacterium, Corynebacterium, Streptococcus, Fusobacterium, uncultured Kiritimatiaellae and Akkermansia. Lactobacillus comprised only 0.18% of the taxonomic composition in estrus and 0.37% in diestrus. No di erences in the relative abundance of the most abundant phylum or genera were observed between estrus and diestrus samples.

The effect of embryo reinsertion immediately after embryo flushing was studied. In Experiment 1, eight mares were used during 32 cycles (8 cycles in each group). For the first two groups, inseminated mares were flushed 8 days after ovulation and prostaglandin F2α was not administered: in group EF-ET (embryo flushing and embryo transfer) the embryo was reinserted in the same donor mare, while in the EF group, no further procedure was performed. In the third group (ET), non-inseminated mares (recipients) received a Day 8 embryo. Progesterone concentration was measured before EF/ET and 72 h after in the three groups. In Experiment 2, twelve mares were used during 17 cycles in two groups, EF-ET (n = 11) and ET (n = 6), as in Experiment 1, except that every mare was flushed 24 h after embryo transfer to retrieve the embryo. Fewer pregnancies resulted after transfer in EF-ET cycles (0/8, 0%) than in the ET group (6/8, 75%). Progesterone concentration decreased significantly (p = 0.05) 72 h after EF-ET but not in EF or ET cycles (p > 0.1). Three mares from the EF-ET showed full luteolysis and signs of endometritis. In Experiment 2, more (5/6; p = 0.08) grade 1 embryos were recovered in the ET compared to the EF-ET group (3/7); 4 embryos were graded 3–4 (were broken or had signs of degeneration) in the EF-ET group but none in the ET group. In both groups, capsule fragments were obtained as indicative of the presence of a recently destroyed embryo in the EF-ET (n = 3) and ET (n = 1) groups. Positive bacterial cultures were obtained in 2/11 and 1/6 embryo flushes from the EF-ET and ET groups, respectively.

Background: PGF2α is commonly given at the end of embryo flushing (EF) to shorten the interval to the next oestrus and ovulation. Objectives: To determine the effect of repeated EF on plasma progesterone concentration, percentage of mares with endometritis, unwanted pregnancy and subsequent fertility in mares flushed without the use of PGF2α. Study design: Controlled experiments. Methods: Nine mares were inseminated in seven consecutive cycles (n = 63), to either perform an EF (n = 54) 7–9 days after ovulation or left pregnant (n = 9). PGF2α was not used to induce oestrus. Ultrasound examination and blood sampling were performed just before the EF and 72 h later to determine changes in progesterone concentration and signs of endometritis. Results: The overall percentage of positive EF/pregnancy was 55.5% (30/54) and 66.7% (6/9), respectively. The likelihood of pregnancy/positive EF in the first three cycles was 55.5% (15/29). This was not different (p > 0.1) from the fertility of the last four cycles (69.4%, 25/36). In five EF cycles (9.3%), mares had signs of endometritis and early luteolysis (progesterone <2 ng/mL) 72 h after EF. The reduction in progesterone concentration by 72 h after EF was greater (p < 0.05) for Day 9 (−2.3 ± 0.7 ng/mL) than Day 7 (−1.0 ± 0.8 ng/mL) or Day 8 (−1.3 ± 1.1 ng/mL) cycles. The progesterone concentration in non-flushed mares did not vary significantly during the sampled period (Day 7–12). There were 5 cycles in which the donor mare remained pregnant after the EF, although four were from a single mare. Main limitations: The mare population was limited to barren and maiden mares. The cycle order and operator allocation to each EF were not randomised. Conclusions: EF induces a subtle, but significant reduction in progesterone concentrations compared with non-EF cycles. However, the percentage of mares with EF-induced full luteolysis is low (9.3%). The fertility of mares after repeated EF without administration of PGF2α was unaffected; however, there is a considerable risk of unwanted pregnancy (5/27 = 18.5%) in donors from which an embryo was not recovered.