During early pregnancy the conceptus and mare communicate to establish pregnancy. Cell-secreted vesicles (e.g. exosomes) have been reported in serum. Exosomes contain bioactive materials, such as miRNA, that can mediate cell responses. We hypothesized that (i) exosomes are present in mare circulation and quantity varies with pregnancy status, (ii) exosomes contain miRNAs unique to pregnancy status, and (iii) miRNAs target pathways in endometrium based upon pregnancy status of the mare. First, serum samples were obtained from mares in a crossover design, with each mare providing samples from a pregnant and non-mated control cycle (n = 3/sample day) on Days 12, 14, 16, and 18 post-ovulation. Flow cytometry revealed the presence of serum microvesicles in mares in two different sized populations (greater than or less than 100 nm), validated by transmission electron microscopy. Second, serum was collected on Days 9, 11, and 13 (n = 4/day) and endometrial biopsies were collected on Days 11 and 13 (n = 3/day) from pregnant and non-mated mares. Total RNA from serum exosomes was evaluated with RT-qPCR using equine-specific miRNA sequences. Twelve miRNAs were found in different quantities at the specified days. Pathway analysis suggested miRNAs targeted focal adhesion molecules. Transcripts corresponding to focal adhesion molecules were evaluated in endometrial biopsies. Protein levels and localization for PAK6 and RAF1 were further evaluated. Our data suggest that serum exosomes contain miRNA that differ based upon pregnancy status and may affect mRNA expression related to focal adhesion pathway in the endometrium with a potential role in maternal recognition of pregnancy.

The mechanism for maternal recognition of pregnancy (MRP) in horses is unknown. To maintain a pregnancy, a mobile conceptus must be recognized by the uterus before d 14 postovulation (PO). This recognition prevents endometrial secretion of PGF_2α on d14 through 16, which would otherwise initiate luteolysis. The objective of this study was to evaluate gene expression in the endometrium of pregnant and nonpregnant mares during and after MRP to identify possible genes involved during this time. Twelve normally cycling mares were used in a crossover design and randomly assigned to a specific collection day. Endometrial samples were collected from a pregnant and nonpregnant (nonmated) mare on cycle d 12, 14, 16, and 18 (n = 3/d) PO. Microarray analysis comparing the endometrial gene expression in pregnant and nonpregnant mares revealed no differences at d 12. Ten genes were identified to have consistently higher or lower expression levels in the endometrium from pregnant versus nonpregnant mares on d 14, 16, and 18 (P < 0.001). The expression of these 10 genes was further analyzed with real-time PCR. d 14, 16, and 18 gene expression patterns were consistent with the microarray analysis, but on d 12, 4 of the 10 were identified as differentially expressed. Endometrial samples were then collected on d 13 PO (n = 3) and processed for western blot and immunohistochemical analysis of 2 proteins due to their reproductive significance. SPLA2 and DKK1 antibody specificity were confirmed via western blot analysis but were not different in samples from pregnant and nonpregnant mares (P = 0.114 and P = 0.514, respectively) and cellular localization was examined by immunohistochemical analysis. This is the first study to describe gene expression and cellular localization in the endometrium at the time of MRP for these genes and suggests that the uterus does not prepare to support a pregnancy until d 14. The function of these genes may be critical in the process of MRP.

Maternal recognition of pregnancy (MRP) in the mare is an unknown process. In a non-pregnant mare on day 14 post-ovulation (PO), prostaglandin F_2α (PGF) is secreted by the endometrium causing regression of the corpus luteum. Prior to day 14, MRP must occur in order to attenuate secretion of PGF. The embryo is mobile throughout the uterus due to uterine contractions from day of entry to day 14. It is unknown what signaling is occurring. Literature stated that infusing oil or placing a glass marble into the equine uterus prolongs luteal lifespan and that in non-pregnant mares, serum exosomes contain miRNA that are targeting the focal adhesion (FA) pathway. The hypothesis of this study is embryo contact with endometrium causes a change in abundance of focal adhesion molecules (FA) in the endometrium leading to decrease in PGF secretion. Mares (n = 3/day) were utilized in a cross-over design with each mare serving as a pregnant and non-pregnant (non-mated) control on days 9 and 11 PO. Mares were randomly assigned to collection day and endometrial samples and embryos were collected on the specified day. Biopsy samples were divided into five pieces, four for culture for 24 hours and one immediately snap frozen. Endometrial biopsies for culture were placed in an incubator with one of four treatments: [1] an embryo in contact on the luminal side of the endometrium, [2] beads in contact on the luminal side of the endometrium, [3] peanut oil in contact on the luminal side of the endometrium or [4] the endometrium by itself. Biopsies and culture medium were frozen for further analysis. RNA and protein were isolated from biopsies for PCR and Western blot analysis for FA. PGF assays were performed on culture medium to determine concentration of PGF. Statistics were performed using SAS (P ≤ 0.05 indicated significance). The presence of beads on day 9 impacted samples from pregnant mares more than non-pregnant mares and had very little impact on day 11. Presence of oil decreased FA in samples from pregnant mares on day 9. On day 11, oil decreased FA abundance in samples from non-pregnant mares. Embryo contact caused multiple changes in RNA and protein abundance in endometrium from both pregnant and non-pregnant mares. The PGF secretion after 24 hours with each treatment was also determined. On day 9, there was no change in PGF secretion compared to any treatments. On day 11, presence of peanut oil increased PGF secretion in samples from non-pregnant mares. In samples from non-pregnant mares, presence of an embryo decreased PGF secretion compared to control samples from non-pregnant mares. Results revealed that while beads and peanut oil may impact abundance of FA RNA and protein in endometrial samples, it does not appear to impact PGF secretion. Conversely, embryo contact for 24 hours with endometrium from a non-pregnant mare causes a decrease in PGF secretion. These results suggest that it is not just contact of any substance/object causing attenuation of PGF secretion, but the embryo itself is necessary to decrease PGF secretion.

Equine maternal recognition of pregnancy (MRP) is a process whose signal remains unknown. During MRP the conceptus and endometrium communicate to attenuate prostaglandin F_2α (PGF) secretion, sparing the corpus luteum and maintaining progesterone production. Recognition of a mobile conceptus by the endometrium is critical by days 14–16 post-ovulation (PO), when endometrium produces PGF, initiating luteolysis. The objective of this study was to evaluate endometrial gene expression changes based upon pregnancy status via RNA sequencing. This experiment utilized a cross-over design with each mare serving as both a pregnant and non-mated control on days nine, 11, and 13 PO (n = 3/status/day). Mares were randomly assigned to collection day and pregnancy confirmed by terminal uterine lavage at the time of endometrial biopsy. Total RNA was isolated and libraries prepared using Illumina TruSeq RNA sample preparation kit. Reads were mapped and annotated using HISAT2 and Stringtie. Expression values were evaluated with DESEQ2 (P ≤ 0.05 indicated significance). On day nine, 11, and 13 there were 1435, 1435 and 916 significant transcripts, respectively. Multiple genes with splice variants had different expression patterns within the same day. These are the first data to evaluate the endometrial transcriptome during MRP on days nine, 11, and 13.