Susceptibility to persistent breeding-induced endometritis and positive embryo flushing are associated with differential endometrial gene expression and uterine extracellular vesicles cargo in mares

Persistent breeding-induced endometritis (PBIE) is the leading cause of subfertility in mares; embryo donor mares are prone to become susceptible to PBIE due to excessive uterine manipulation after repeated breeding-flushing cycles, coupled with age-related deterioration of the cervix and uterus. Therefore, the study's objectives were to determine: (i) if a differential endometrial response to the embryo exists (i.e., positive embryo flush) on Day-8 post-ovulation and (ii) if this endometrial response is different in mares susceptible to PBIE. Twenty mares were screened for susceptibility to PBIE using a two-billion-frozen-killed sperm model coupled with Kenney & Doig's scoring; and, based on their responses, classified as susceptible (n=6) or resistant (n=6) to PBIE or intermediate (n=8). Mares were bred with two-billion sperm from one fertile stallion. On Day-8 post-ovulation, mares were submitted to low-volume uterine lavages (i.e., 100 mL), followed by embryo flushing (4 L) and collection of endometrial biopsies. Two embryos were collected from susceptible and three from resistant mares; seven breeding cycles yielding a negative embryo flushing were included as controls. Endometrial luminal (LE) and glandular epithelium (GE) were micro-dissected with ZEISS PALM Microbeam and analyzed by RNA-sequencing. Uterine extracellular vesicles (uEVs) were isolated from uterine fluid, and their RNA content was analyzed. The comparison betweenpositive- and negative embryo-flushing revealed 305 differentially expressed genes (DEGs) for LE (FDR 5%) while 50 DEGs for GE (FDR 20%, P<0.001), indicating that the main response is present in the LE. The most significantly overrepresented functional terms for genes upregulated in LE on Day-8 were related to ‘response to hormones,’ ‘lipid or extracellular stimulus,’ ‘cell cycle,’ and ‘T-cell activation.’ The analysis of uEVs revealed 442 differentially abundant (DA) RNAs (FDR 5%) between mares with positive- and negative-embryo flushing. At the miRNA level, miR-500-5p and miR-1246 were found in higher concentrations in uEVs from mares yielding positive-embryo flushing. Overrepresented functional terms were related to, e.g., ‘translation,’ ‘lipid binding,’ ‘cell cycle, ‘cytokine binding,’ and ‘angiogenesis.’ The comparison between susceptible and resistant mares resulted in 80 DEGs for GE (FDR 20%, P<0.001) and 65 DEGs for LE (FDR 28%, P<0.001). Functional analysis revealed enrichment for: ‘lipid metabolism and biosynthesis,’ ‘response to the hormone,’ ‘RNA splicing,’ and ‘cell morphogenesis.’ Analysis of uEVs revealed 7 DA RNAs (FDR 20%) and one DA miRNA (eca-miR-181a-5p) between susceptible- and resistant-mares, which has been described as associated with fertility. In conclusion, this is the first study showing a transcriptional response to the presence of an embryo on Day-8 in terms of alterations in endometrium and uEVs cargo and revealing differences between mares susceptible and resistant to PBIE.