Prostaglandin-related genes are differentially expressed in equine endometrium with different biopsy grade, degrees of inflammation, and fibrosis.

Prostaglandins have many roles in the equine reproductive tract, including but not limited to luteolysis, luteal support, ovulation, transport through the uterine tube, uterine contraction, embryonic mobility, inflammation, and fibrosis. Altered secretion of inflammatory proteins are likely to disrupt the balance of endometrial function and could impair fertility. Our overall goal was to measure the expression of several prostaglandin- and inflammation-related genes in mares with different degrees of endometrial histological changes. Our hypothesis was that mares with neutrophilic and lymphocytic plasmocytic inflammation, fibrosis, or different biopsy grades would have altered concentrations of prostaglandin E2 (PGE2) and F2α (PGF2α), as well as altered expression of inflammation- and prostaglandin-related genes, compared to mares with minimal to no histological changes on biopsy evaluation. Forty-five endometrial biopsies from estrous mares were assessed by a reproductive pathologist for the degree of neutrophilic inflammation, lymphocytic and plasmocytic inflammation, and fibrosis, and a biopsy grade was assigned based on the Kenney-Doig system. A low-volume uterine lavage was collected from a subset of twenty-six mares prior to biopsy collection and was used to measure PGE2 and PGF2α concentrations via ELISA. Total RNA was extracted from biopsies and mRNA expression was evaluated for twenty-five genes of interest. A restricted maximum likelihood linear model was used to compare differences of mRNA expression, with a statistical significance set at P < 0.05. There was no difference in the abundance of PGE2 or PGF2α between any of the variables tested. Mares with endometrial biopsy grade I had lower expression of NF-kB, PTGS1 and HPGD compared to grade IIA or IIB (P < 0.05). Mares with neutrophilic inflammation had decreased expression of NF-kB, PTGS1, PTGER4, CBR1, mPGES2 and PTGIS compared to mares without inflammation. Mares with mild or minimal endometrial fibrosis had increased expression of mPGES2 and PTGIS, compared to mares with moderate endometrial fibrosis. In conclusion, several genes were identified to be differentially expressed in mares with histological changes compared to mares with no to minimal histological changes. The presence of inflammation and fibrosis may alter the concentration of prostaglandins in endometrial tissue, which could impair many of the uterine reproductive and immune functions during estrus, affecting early embryo survival.

Inflammatory markers for differentiation of endometritis in the mare.

BACKGROUND: Endometrial biopsy is required to diagnose mares with chronic endometritis and endometrial degenerative fibrosis. An increase in understanding of equine reproductive immunology could be utilised to create less-invasive, time-efficient diagnostic tools especially when evaluating mares for chronic endometritis. OBJECTIVES: To evaluate inflammatory cytokine and chemokine concentrations in uterine fluid samples collected by low-volume lavage (LVL) as a potential screening diagnostic biomarker for endometritis. STUDY DESIGN: Prospective cross-sectional clinical study. METHODS: Forty-six mares underwent a LVL and subsequently endometrial biopsy. Mares were split in three groups: healthy, acute endometritis, and chronic endometrial fibrosis (CEF) based on cytological and histological evaluation. A fluorescent bead-based multiplex assay for IFN-γ, IFN-α, IL-1ß, IL-4, IL-10, IL-17, sCD14, TNF-α, CCL2, CCL3, CCL5 and CCL11 were carried out on the LVL fluid. The endometrial biopsy was utilised for histology and qPCR of IFN-γ, IL-1ß, IL-6, IL-8, IL-17, TNF-α, CCL2 and CCL3 genes. Statistical analyses examined differences in inflammatory markers and predictive modelling for diseased endometrium. RESULTS: Secreted concentrations of IFN-γ were lower in LVL fluid from reproductively healthy mares compared with acute endometritis (p = 0.04) and CEF (p = 0.006). Additionally, IL-17, IL-10, IL-1ß, TNF-α, CCL2, CCL3, CCL5 and CCL11 were significantly increased (p ≤ 0.04) in LVL from CEF mares compared with healthy mares. Mares with CCL2 concentrations ≥550 pg/mL (14/14) had 100% probability of having CEF and/or acute endometritis. Healthy mares had lower relative abundance of IL-17 mRNA compared with mares in CEF group [median (interquartile rage) = 14.76 (13.3, 15.3) and 12.4 (10.54, 13.81)], respectively (p = 0.02). MAIN LIMITATIONS: Limited sample size: larger numbers of mares with and without endometritis are required and reference intervals in LVL samples have to be established. CONCLUSIONS: Inflammatory chemokines and cytokines concentrations differed between healthy mares and mares with acute endometritis or CEF in LVL.

Characterization of serum and tissue oxytocinase and tissue oxytocin in the pregnant and non-pregnant mare.

Oxytocin is a hormone with functions in: reproduction, maternal bonding, milk ejection, and feeding/social behavior, and is reported to be present in a variety of tissues. Our goal is to characterize oxytocin and leucyl and cystinyl aminopeptidase (LNPEP/oxytocinase), a key regulator of oxytocin in mares. We measured serum and tissue LNPEP by ELISA from ovulation (D0) until D21-22 in non-pregnant (n = 5) and pregnant mares (n = 6); and in periparturient and postpartum mares (n = 18). Placenta (n = 7) and homogenized tissue of diestrus mares (n = 6) were evaluated using protein determinations and LNPEP ELISAs. Identification of LNPEP and OXT protein in tissues was also performed via western blot, immunohistochemistry and liquid chromatography-mass spectrometry (LC-MS/MS). Furthermore, in situ hybridization was performed for LNPEP and OXT on endometrium, myometrium, pituitary and corpus luteum (CL). Serum LNPEP concentration were similar. Placental LNPEP U/mg protein was highest in the body and pregnant horn. The highest to lowest LNPEP U/mg protein by tissue were: myometrium > follicle wall > endometrium > kidney > CL > liver. Oxytocin was identified in the equine pituitary, CL and placenta and is likely to act in autocrine or paracrine manner, while LNPEP may act systemically and locally to regulate the availability of OXT.

Serum progesterone and oxytocinase, and endometrial and luteal gene expression in pregnant, nonpregnant, oxytocin, carbetocin and meclofenamic acid treated mares.

Our objectives were to examine changes in endometrial and luteal gene expression during estrus, diestrus, pregnancy and treatments to induce luteolysis and putatively induce luteostasis. Groups were: Diestrus (DIEST), Estrus (ESTR), Pregnant (PREG), Oxytocin (OXY), Carbetocin (CARB), and Meclofenamic acid (MFA). Blood was obtained from day (D)12 to D15 for measurement of oxytocinase, also referred to as leucyl-cysteinyl aminopeptidase (LNPEP) and progesterone. Luteal biopsies were obtained on D12 and D15 and an endometrial biopsy on D15. Real-time RT-PCR was performed for the following genes: PGR, ESR1, OXTR,OXT, LNPEP, PTGS2, PTGFR, PLA2G2C, PTGES, SLC2A4, and SLC2A1. Regarding serum LNPEP, PREG and OXY (p-value<0.001) had higher concentrations than DIEST mares. Endometrial PTGES expression was higher (p-value <0.04) in DIEST, PREG and OXY than other groups. Endometrium from ESTR had increased expression of OXT (p-value < 0.02) compared to MFA and OXY mares. Carbetocin treatment: decreased serum progesterone and LNPEP; increased endometrial PLA2G2C; decreased endometrial PTGES; and decreased luteal aromatase and PTGES. Treatment with MFA: decreased endometrial PLA2G2C, increased endometrial PTGES; and resulted in less OXTR and OXT luteal abundance on D12 compared to D15. Endometrial and luteal expression of LNPEP is affected by physiologic stage and treatment and is involved in luteal function and pregnancy recognition pathways through effects on oxytocin and prostaglandin synthesis in the horse.

Endometrial and luteal gene expression of putative gene regulators of the equine maternal recognition of pregnancy.

Our understanding of the temporal changes in endometrial and luteal gene transcripts related to the actions of oxytocin and prostaglandin during early equine pregnancy is incomplete. Additionally, the role of oxytocinase, also known as Leucyl-cystinyl aminopeptidase (LNPEP), during early pregnancy in mares has not been previously investigated. Luteal and endometrial biopsies were obtained on Day (D)8, D10, D12 and D15 post-ovulation in pregnant (PREG) and diestrus (DIEST) mares for real-time qPCR. Differences in endometrial gene expression occurred over time in: SLC2A4, SLC2A1, PTGES, OXTR and LNPEP. PTGFR and PLA2G2C had lower relative abundance in PREG D15 endometrium compared to D10. OXT and OXTR were increased on D10 and 15 PREG, respectively. Regarding luteal mRNA relative abundance, ESR1, PTGS2, PTGFR, and PTGES had higher relative abundance in D12 of DIEST and PREG. Luteal expression of OXTR and OXT had higher relative abundance in D15 compared to D8, and LNPEP had higher relative abundance in D10 and 12. Endometrial and luteal PTGES had an increased mRNA abundance in both D12 DIEST and PREG mares, which may lead to additional luteoprotective prostaglandin E2 (PGE2) secretion. Furthermore, luteal SLC2A1 had higher relative abundance in pregnancy, and likely supports the high metabolic activity of luteal tissue by increasing glucose uptake. Oxytocinase is present in endometrial and luteal tissue and its role in oxytocin induced prostaglandin secretion is uncertain.

Expression of Oxytocin/Neurophysin I and Oxytocinase in the Equine Conceptus from Day 8 to Day 21 Post-Ovulation.

Leucyl and cystinyl aminopeptidase (LNPEP/oxytocinase) is an enzyme that metabolizes oxytocin in serum and tissues. The presence of oxytocin/neurophysin I (OXT), oxytocin and LNPEP and their relationship to other genes is unknown in the equine conceptus. Our objective was to characterize gene expression of LNPEP and OXT on D8, 10, 12, 14, 15, 16 and 21 conceptuses in relationship to other genes. Immunohistochemistry, western blot and liquid chromatography with tandem mass spectrometry (LC-MS/MS) were used for identification of oxytocin and LNPEP in D15, 16 and 18 conceptuses. LNPEP was increased at D15 compared to D10, was immunolocalized in the equine trophectoderm and endoderm, and protein was confirmed by LC-MS/MS. Maximal abundance of OXT was at D21, and lowest on D12 and D14, but no protein was identified. OXTR abundance was highest on D14 and D21. LNPEP was correlated with PTGFR and PTGES on D12 and D14-D15, and high expression of PTGES, PTGS2 was found on D14, D15 and D21; PTGFR was found on D8 and D12-21. LNPEP may have a role in prostaglandin regulation and conceptus fixation by decreasing the availability of oxytocin. Further investigation on the role embryonic LNPEP during pregnancy is warranted.

Analysis of the uterine flush fluid proteome of healthy mares and mares with endometritis or fibrotic endometrial degeneration.

The objective of this study was to evaluate the differences in the uterine flush fluid proteome between healthy mares and mares with endometritis or fibrotic endometrial degeneration (FED). Uterine flush fluid samples were collected from healthy mares (n=8; oestrus n=5 and dioestrus n=3) and mares with endometritis (n=23; oestrus n=14 and dioestrus n=9) or FED (n=7; oestrus n=6 and dioestrus n=1). Proteomic analysis was performed using label-free liquid chromatography-tandem mass spectrometry. Of 216 proteins identified during oestrus, 127 were common to all three groups, one protein was exclusively detected in healthy mares, 47 proteins were exclusively detected in mares with endometritis and four proteins were exclusively detected in mares with FED. Of 188 proteins identified during dioestrus, 113 proteins were common between healthy mares and mares with endometritis, eight proteins were exclusively detected in healthy mares and 67 proteins were exclusively detected in mares with endometritis. Quantitative analysis revealed a subset of proteins differing in abundance between the three groups during oestrus and between healthy mares and mares with endometritis during dioestrus. These results provide a springboard for evaluation of specific proteins as biomarkers of uterine health and disease and for investigation of their roles in the establishment and maintenance of pregnancy.

Impact of decision to delivery time of fetal mortality in canine caesarean section in a referral population.

In human medicine, there is a recommended decision to delivery interval (DDI), which allows for the optimization of protocols and systematic review of hospital success. In veterinary medicine, no such guideline has been established or investigated. The purpose of this study was to investigate the relationship between the interval from the decision to perform a caesarean section and the delivery of the neonates and fetal mortality at the time of surgery. One hundred and fifty canine caesarean sections were evaluated retrospectively. Caesarean cases were dichotomized to those that had at least one perinatal death and cases where all puppies survived. Factors that increased the likelihood of at least one perinatal death at caesarean section were: cases that presented as an emergency caesarean section, the dam presenting with a fetus in the vaginal canal, the dam not having a history of previous caesareans sections, and being multiparous. Even though there was no association of DDI with likelihood of having perinatal death, timing related factors that increased the likelihood of having at least one perinatal death at caesarean section were: cases where total anaesthesia time was longer than 2 h; time from induction to start of surgery was longer than 45 min and surgical time longer than 75 min. In conclusion, time is a factor in the success of canine caesarean sections and further research is needed to better define the optimal decision to delivery time in canine caesarean sections in order to optimize fetal survival and hospital protocol success rates.

A comparison of the uterine proteome of mares in oestrus and dioestrus.

Proteomic analysis of mare uterine flush fluid provides a minimally invasive technique for studying protein changes associated with the oestrous cycle. The aim of this study was to identify differentially abundant proteins in the uterine flush fluid of mares in oestrus and dioestrus. In this study, uterine flush fluid samples were collected from eight reproductively healthy mares in either oestrus (n = 5) or dioestrus (n = 3). Proteomic analysis was performed using liquid chromatography-tandem mass spectrometry. Of 172 proteins identified, six proteins (immunoglobulin lambda-like polypeptide 1, haemoglobin subunit alpha, alpha-1B-glycoprotein, serotransferrin, apolipoprotein A-1, and haemoglobin subunit beta) were significantly more abundant in oestrus. These proteins may contribute to the endometrial defence system through roles in inflammation, immunity or antimicrobial activity. In other species, some of these proteins have been described as immunoglobulins, negative acute phase proteins or defence agents against micro-organisms. During dioestrus, immunoglobulin alpha-1 chain C region-related, complement factor I, CD 109 antigen and uterocalin, were significantly more abundant. Research in other species suggests that these four proteins contribute to the immune response through proposed immunoregulatory characteristics, complement system involvement or roles in B cell-T cell interactions. In conclusion, ten differentially abundant proteins were identified in the uterine flush fluid of mares in oestrus and dioestrus. Targeted studies on these proteins could elucidate their role in uterine defence mechanisms during the oestrous cycle in the mare.

Progesterone levels and days to luteolysis in mares treated with intrauterine fractionated coconut oil.

Intrauterine plant oil infusion, including fractionated coconut oil, has been previously found to be a safe, inexpensive, and reversible method of prolonging the luteal phase in mares when administered on Day 10 of the estrous cycle. Our objective was to understand the uteroovarian response to the administration of fractionated coconut oil infusion in the uterus of diestrous mares. We hypothesized that intrauterine coconut oil administration on Day 10 would prolong luteal life span in a dose-dependent fashion and would result in higher serum progesterone levels than untreated mares at the expected time of luteolysis. Light-horse mares (n = 18) were examined using transrectal palpation and ultrasonography to determine if they had a normal interovulatory interval and were then examined daily in estrus until the day of ovulation (Day 0) and then every other day during an estrous cycle. Jugular blood was drawn on Day 11, Day 13, Day 15, and Day 17, centrifuged, and serum stored until assayed for progesterone (P4; Siemens Coat-a-Count Progesterone RIA, Los Angeles, CA, USA). Mares were randomly assigned to treatment and studied over one to two estrous cycles with a rest cycle after each treatment cycle. Groups were: control (n = 5), fractionated coconut oil 1.0 mL (Miglyol 810; Sasol Oil, Witten, Germany) infused in the uterus with an artificial insemination pipette on Day 10 (Group 1; n = 5) and fractionated coconut oil 0.5 mL infused in the uterus with an embryo transfer gun, on Day 10 (Group 2; n = 5). All statistical analyses were performed using analytical software (Stata SE, version 13.1, College Station, TX, USA) at P < 0.05. Data were first evaluated using the Shapiro-Wilk test for normality. Differences between groups in days to luteolysis (DTL) were examined using analysis of variance and Bonferroni, and the effect of day and treatment on P4 levels were examined using the Kruskal-Wallis and Dunn's all pairwise test. There was a significant difference in DTL between the groups (P = 0.0083), with fewer DTL in Group 1 compared to control (P = 0.011) and to Group 2 (P = 0.034). There was a significant effect of day (P < 0.0001) on P4 levels with Day 11 P4 levels higher than Day 15 and Day 17 and Day 13 P4 levels higher than Day 15 or Day 17 (P < 0.03). There was a significant effect of treatment group (P = 0.0098) on P4 levels with control and Group 2 levels higher than Group 1 mares (P = 0.0012, P = 0.0495, respectively). We concluded that intrauterine administration of 1 mL of fractionated coconut oil lowered P4 levels in diestrus in a dose-dependent fashion and did not prolonged the luteal phase of the mares.

Evaluation of diagnostic utility, safety considerations, and effect on fertility of transvaginal ultrasound-guided ovarian biopsy in mares.

Ultrasound-guided biopsies of corpora lutea have been previously used for research purposes in the mare and cow. However, the health effects and fertility after transvaginal luteal biopsies (TVLB) or transvaginal ovarian biopsies (TVOB) obtained for diagnostic purposes in cases of suspected ovarian tumors have not been previously evaluated in the horse. The aim of this study was to determine the effects on health and fertility of TVLB and TVOB in mares; 53 mares were included in the study (11 control non-biopsied mares, 37 TVLB mares biopsied on one or more of the following Days 8, 10, 12, 15, 21, and 5 TVOB mares with ovarian abnormalities), resulting in a total of 108 TVLB and TVOB cycles and 183 procedures. Mares were divided into Groups 1 to 3 by the number of TVLB per estrous cycle (mare in Group 1 had 1 TVLB procedure, mares in Group 2 had 2 TVLB procedures, and mares in Group 3 had 3 TVLB procedures). Group 4 comprised TVOB mare cycles with ovarian abnormalities (n = 5). Mares were examined to determine day of ovulation (Day 0) and the presence of a corpora lutea using transrectal ultrasonography. Mares were sedated, and an ultrasound-guided transvaginal biopsy was performed of luteal or ovarian tissue. Health effects of TVLB or TVOB were assessed by daily rectal temperatures, appetite, and general demeanor for 72 hours post-procedure, and 3 mares were examined at necropsy. Fertility was not significantly different in control and TVLB Groups 1 to 3 (P = 0.7648) and in the first or subsequent cycles where the ovulation was from an ovary that had a previous TVLB (P = 0.7147). A TVLB on Day 8 post-ovulation may induce an early return to estrus. In conclusion, the TVLB or TVOB procedure had no effect on health and fertility in this study if the procedure was correctly performed with good technique. Because of the low number of cycles (n = 37), the fertility data should be interpreted with caution. The TVOB may be applied in the diagnosis of mares with ovarian abnormalities, and no adverse health effects were associated with TVOB of mares with granulosa theca cell tumor.