A mathematical model of in vivo bovine blastocyst developmental to gestational Day 15.

Bovine embryo growth involves a complex interaction between the developing embryo and the growth-promoting potential of the uterine environment. We have previously established links between embryonic factors (embryo stage, embryo gene expression), maternal factors (progesterone, body condition score), and embryonic growth to 8 d after bulk transfer of Day 7 in vitro-produced blastocysts. In this study we recovered blastocysts on Days 7 and 15 after artificial insemination to test the hypothesis that in vivo and in vitro embryos follow a similar growth program. We conducted our study using 4 commercial farms and repeated our study over 2 yr (2014, 2015), with data available from 2 of the 4 farms in the second year. Morphological and gene expression measurements (196 candidate genes) of the Day 7 embryos were measured and the progesterone concentration of the cows were measured throughout the reproductive cycle as a reflection of the state of the uterine environment. These data were also used to assess the interaction between the uterine environment and the developing embryo and to examine how well Day 7 embryo stage can be predicted from the Day 7 gene expression profile. Progesterone was not a strong predictor of in vivo embryo growth to Day 15. This contrasts with a range of Day 7 embryo transfer studies which demonstrated that progesterone is a very good predictor of embryo growth to Day 15. Our analysis demonstrates that in vivo embryos are 3 times less sensitive to progesterone than in vitro-transferred embryos (up to Day 15). This highlights that caution must be applied when extrapolating the results of in vitro embryo transfer studies to the in vivo situation. The similar variance in measured and predicted (based on Day 15 length) Day 7 embryo stage indicate low stochastic perturbations for in vivo embryo growth (large stochastic growth effects would generate a significantly larger standard deviation in measured embryo length on Day 15). We also identified that Day 7 embryo stage could be predicted based on the Day 7 gene expression profile (58% overall success rate for classification of 5 embryo stages). Our analysis also associated genes with each developmental stage and demonstrates the high level of temporal regulation of genes that occurs during early embryonic development.

Influence of pathogenic bacteria species present in the postpartum bovine uterus on proteome profiles.

In the first 2-3 weeks after parturition >90% of dairy cows will have some form of uterine infection. Uterine contamination with pathogens, such as Trueperella (formerly Arcanobacterium) pyogenes increases the risk of developing more severe endometritis, which can reduce conception rates. In this study, we compared the uterine proteome of cows infected with Trueperella pyogenes with that of uninfected cows, using 2D gel electrophoresis, and identified annexins A1 and A2 (ANXA1 and ANXA2), apolipoprotein A-1, calprotectin (S100A9), cathelicidin, enolase 1 (ENO1), peptidoglycan recognition protein 1 (PGLYRP1), phosphoglycerate mutase 1 (PGAM1), serine dehydratase (SDS) and serine protease inhibitors (SERPIN) B1, B3 and B4 proteins as differing in abundance in endometritis. Subsequently, levels of ten of these proteins were monitored in uterine samples collected from a herd of lactating, dairy cows at 15 and 42 days post-partum (DPP). The levels were compared with the cytology scores of the samples and the bacterial species isolated from the uterus. Cathelicidin, PGLYRP1, SERPINB1 and S100A9 levels at 15DPP showed strong positive correlations (r=0.78, 0.80, 0.79, and 0.68 respectively; P<0.001) with % of polymorphonuclear neutrophils (PMN). When compared with other bacterial pathogens identified, Streptococcus agalactiae and Truperella pyogenes induced increased expression of the indicator proteins, suggesting that these organisms may adversely affect the subsequent ability of the cow to conceive. Interestingly, there was no difference in the proportion of cows pregnant at 6 and 17 weeks after start of mating between the cows with high or low %PMN.

Effect of asynchronous transfer on bovine embryonic development and relationship with early cycle uterine proteome profiles.

The uterus provides the nurturing environment that supports the growth of the early preimplantation bovine conceptus. To determine critical time points of uterine influence, in vitro-produced Day 7 blastocysts were transferred into synchronous (Day 7) uteri and asynchronous uteri (Days 5 or 9). Embryo growth was evaluated 7 and 15 days after transfer and compared with that of embryos generated by AI. Conceptuses recovered from asynchronous Day 9 transfers were fourfold larger than synchronous transfer or gestational Day 14 AI conceptuses; by 15 days after transfer, differences were less marked. Two-dimensional gel electrophoresis was used to compare the histotroph protein composition of uterine luminal flushings (ULF) on Days 5 and 9 after oestrous to determine any protein differences that would promote embryo growth. The ULF were collected by serially flushing the uteri of the same heifers and mature cows at different times of the cycle. Ten proteins that differed in abundance between Day 5 and 9 were identified by mass spectrometry. Three, namely phosphoserine aminotransferase 1, purine nucleoside phosphorylase and aldose reductase, were verified by western blot analysis as more abundant on Day 9 (P<0.002). Myostatin was present in only in Day 9 ULF, whereas tissue inhibitor of matrix metalloproteinase 2 (TIMP2) and legumain were only detected in Day 14 ULF. Although mature cows had lower progesterone concentrations on Days 5 and 14 (P<0.05) and tended to have less TIMP2 than heifer groups, no other protein differences were detected. Thus, the embryo growth-enhancing environment on Day 9 was associated with temporal changes in the expression of several proteins of the histotroph.

Evaluation of the uterine environment early in pregnancy establishment to characterise cows with a potentially superior ability to support conceptus survival.

During previous investigations, the capacity of the cow to secrete prostaglandin in response to oxytocin has been linked with pregnancy outcome. The objective of the present study was to evaluate the predictive value of prostaglandin release to identify groups of cows as potentially superior (SR, low prostaglandin release) or inferior (IR, high prostaglandin release) for pregnancy outcome and to utilise these cows to investigate factors that contribute to optimum uterine conditions for early pregnancy. Animals were synchronised and received an in vitro-derived blastocyst on Day 7 post-oestrus. Tissues (trophoblast and endometrial) and uterine luminal fluid (ULF) were recovered 10 days later. Pregnancy rates were 94 and 78% for SR and IR cows, respectively. Of the pregnant SR cows, 69% had larged conceptuses (>24 cm) in contrast to 43% IR of cows. IR cows with small conceptuses (<12 cm) had significantly lower mean Day 3 and 5 post-oestrous progesterone concentrations than cows with large conceptuses. The expression of factors involved in the prostaglandin pathway, pregnancy and conceptus development were analysed via quantitative RT-PCR and Western blot analysis. Investigation of 16 endometrial gene transcripts indicated no differences between IR and SR cows except for osteopontin expression which, in uteri with large conceptuses, was 2-fold greater in SR than IR cows (P=0.02). There was greater expression of CTGF, OXTR, PGES, PGHS2 and UTMP mRNA in uteri of SR and IR cows that had large compared with small conceptuses (P<0.05). More IFNT protein was recovered in SR compared with IR ULF (P<0.03). SR cows with large conceptuses had less TIMP2 and legumain protein in their gravid, compared with their non-gravid horns (P≤0.02) whereas IR cows did not. The predictive value of prostaglandin release in response to oxytocin challenge does not appear to be an effective indicator of subsequent pregnancy rates in cows. Differences between the two groups appear to be associated with subtle differences in progesterone and uterine protein concentrations that may be related to differences in conceptus size.

Long-term alteration of follicular steroid concentrations in relation to subclinical endometritis in postpartum dairy cows.

The focus of this study was to investigate the effect of subclinical endometritis (scEndo) on ovarian follicular steroid concentrations in early postpartum pasture-fed dairy cows. Mixed-age lactating dairy cows (n = 169) were examined to ascertain uterine health status on d 21 postpartum (±3 d). From this herd, a cohort of scEndo and uninfected cows (n = 47) were selected using uterine cytology to determine scEndo. To ensure cows with scEndo were selected for the study, a conservative threshold [>18% polymorphonuclear (PMN) cells among uterine nucleated cells] was chosen as a selection threshold. Ovarian follicular dynamics were assessed by ultrasonography on d 21, 42, and 63 postpartum. On the latter 2 d, all follicles >4 mm in diameter were ablated, and 4 d later, the largest (F1) and second largest (F2) follicles were measured and their follicular fluid aspirated. Hematological variables and plasma metabolites were measured also on these days to further characterize scEndo cows. On d 21, the prevalence of scEndo was approximately 9% in this herd; by d 42 infections had self-resolved in the majority (81%) of those cows classified as having scEndo on d 21. The scEndo cows had a delayed return to cyclicity; however, no effect was evident on ovarian follicle size or growth rate. Weeks after scEndo had self-resolved and cyclicity was restored, decreased (P = 0.07) testosterone and increased (P = 0.07) cortisol concentrations were evident in F1 follicles of scEndo compared with uninfected cows. Progesterone concentrations of F1 increased (P < 0.05) in 11- to 16-mm diameter follicles of scEndo cows, whereas estradiol, androstendione, and dehydroepiandrosterone concentrations were decreased (P < 0.05) in F1 8- to 10-mm diameter follicles of scEndo cows. These 3 steroids also differed (P < 0.05) between F1 follicle size categories of scEndo but not uninfected cows. On d 21, mean plasma albumin concentration was decreased (P = 0.02) in scEndo cows. In summary, early postpartum scEndo had surprisingly long-term influences on the steroid concentrations of ovarian follicles long after infections had self-resolved. This is likely to affect oocyte quality and may partially explain the reduced conception rates and longer interval between calving and conception that are often associated with scEndo, although more detailed investigations are required to substantiate this theory.

Polyunsaturated fatty acids differentially alter PGF(2alpha) and PGE(2) release from bovine trophoblast and endometrial tissues during short-term culture.

Two studies tested the hypothesis that eicosapentaenoic (20:5omega3; EPA), docosahexaenoic acids (22:6omega3; DHA) or linoleic acid (C18:2omega6; LIN) reduced bovine endometrial and trophoblast prostaglandin F(2alpha) (PGF(2alpha)) and prostaglandin E(2) (PGE(2)) release during short-term culture. In Study 1, endometrial tissues were collected from non-lactating, non-pregnant cows and endometrial plus trophoblast tissues from pregnant cows 16 days post-insemination. In Study 2, endometrial and trophoblast tissues were collected on day 17 of pregnancy, from cows synchronised using a double prostaglandin (PG) or Ovagentrade mark synchronisation. Tissues were incubated in medium only (M) or media supplemented with fatty acids: eicosapentaenoic (20:5omega3; EPA), docosahexaenoic acids (22:6omega3; DHA) or linoleic acid (C18:2omega6; LIN). In Study 1, PGE(2) release from 'pregnant' endometria was higher (P=0.094) than from 'non-pregnant' endometria, while PGF(2alpha) concentrations were similar. Fatty acids treatment had no effect on PGF(2alpha) or PGE(2) release from either pregnant or non-pregnant endometria. Individual fatty acid treatments had no effect on the ratio of PGF(2alpha) to PGE(2) from trophoblast tissues, but when the data from the 3 fatty acid treatments were combined (EPA, DHA and LIN treatment groups) the ratio of PGF(2alpha) to PGE(2) was reduced (P=0.026) when compared to medium only. In Study 2, PGE(2) concentrations were higher (P=0.013) from the trophoblast collected from Ovagentrade mark cows as compared to that of the PG synchrony group. When the data from the 3-omega fatty acids were combined (DHA and EPA treatment groups), the 3-omega treatments decreased (P<0.05) PGE(2) biosynthesis from both endometrial and trophoblast tissues from animals synchronised following PG synchrony but not Ovagentrade mark synchrony. Short-term culture with low concentrations of 3-omega fatty acids tended to reduce prostaglandin release from trophoblast collected 16 days after insemination, with the type of synchrony modifying PGE(2) production from the trophoblast tissues collected 17 days after insemination. The ability of exogenous fatty acids to modify embryonic prostaglandin release needs to be examined in the context of supplementing dairy cows with different sources of fats. Synchronisation method altered trophoblast PGE(2) release, highlighting the importance of the hormonal environment in modifying embryonic prostaglandin synthesis and release.

Bovine endometrial legumain and TIMP-2 regulation in response to presence of a conceptus.

The precise mechanism of placentation in the bovine species where a restricted trophoblast invasion occurs to form the synepitheliochorial placenta is not fully understood. This study initially investigated the conceptus-maternal interactions in the peri-attachment period by comparing the proteins present at Days 16 and 18 in uterine luminal fluid (ULF) of pregnant with nonpregnant cows using 2-D gel electrophoresis. Nine protein spots were identified that were present in greater amounts in pregnant compared to nonpregnant ULF: carbonic anhydrase, ezrin, heat shock protein 70, isocitrate dehydrogenase, nucleoside diphosphate kinase, peroxiredoxin 1, purine nucleoside phosphorylase, thioredoxin and triosephosphate isomerase and four proteins that were less abundant in ULF from the gravid compared to the nongravid horns or nonpregnant uteri: cystatin E/M, legumain, retinol-binding protein (RBP) and tissue inhibitor of matrix metalloproteinase 2 (TIMP-2). Successful placentation requires the remodelling of the endometrial surface therefore uterine mRNA and protein expression of legumain, a protease activator, and TIMP-2, a protease inhibitor, was examined in detail during the oestrous cycle and from Days 13 to 31 of pregnancy. Both mRNAs were up-regulated in the endometrium during the luteal phase of the oestrous cycle and during early pregnancy. Although legumain and TIMP-2 mRNA expression levels were similar between uterine horns at the same day of pregnancy, the amount of protein differed between gravid and nongravid horns possibly modulated by interferon-tau or by other factors produced by the conceptus. These events at the conceptus-maternal interface may provide localised control of protease activity necessary for controlling trophoblast invasion of the endometrium.

Expression of genes associated with allantois emergence in ovine and bovine conceptuses.

In the development of ruminant embryos, the emergence and growth of the allantois is critical for the establishment of the chorioallantoic placenta. The allantoic membrane contributes to all the vasculature that perfuses the placental tissues and the fetal membranes. Using suppressive subtractive hybridization to compare mRNA from Day 13 ovine preimplantation conceptuses (prior to allantoic emergence) with Day 17 allantoic membrane, we identified nine genes whose expression was associated with the emergence of the allantoic sac. Collagen alpha 1 type XII, collagen alpha 2 type I, collagen alpha 2 type V, epsilon 4 beta-globin, osteonectin, and uroplakin were expressed at significantly greater levels in ovine Day 17 allantois compared to Day 13 conceptuses. These genes are associated with the extracellular matrix and most likely are involved in establishing and strengthening the structural integrity of the allantoic sac and in the development of the blood vessels. RalB expression increased with development although at significantly greater levels in the allantois only at Day 19. Hoxa-10 and RhoA showed no differential expression during this period. All these genes showed a similar temporal pattern of expression in bovine conceptuses at equivalent stages of development with significantly greater expression of all these genes, except for Hoxa-10, found in Day 24 allantois compared to Day 14 conceptuses. This suggests that the role they play in allantoic emergence, growth and function is conserved in both ruminant species and that their expression is regulated in a similar manner. The interactions and regulation of this process remains to be fully explained.

Oestrogen regulation of insulin-like growth factor binding protein-3 (IGFBP-3) and expression of IGFBP-3 messenger RNA in the ovine endometrium.

Transcriptional regulation of insulin-like growth factor binding protein-3 (IGFBP-3) by oestrogen was investigated by Northern blot hybridization of endometrial tissue mRNA from anoestrous ewes treated with oestradiol-17beta at either 0, 40 or 80 microg per day for 7 days. The 2.6 kb IGFBP-3 transcript, seen in control animals, was virtually undetectable in treated animals. This suppressive effect was reflected in Western ligand blots of the uterine luminal fluid (ULF) proteins where the concentration of IGFBP-3 was significantly decreased with increasing oestrogen treatments. IGFBP-2 levels were increased with oestradiol treatment but no significant effect was seen on the other minor IGFBP's present in the ULF. Northern analysis also showed that the IGFBP-3 transcript was present from days 12 to 16 of the oestrous cycle but was either absent or very weak on days 0 (oestrus) and 9 of cycling ewes. In situ hybridization of endometrial tissue sections localized the IGFBP-3 mRNA to the luminal epithelial cell layer, areas of the stromal tissue and in some glandular epithelial cells. Oestradiol treatment of ewes down-regulated expression of IGFBP-3 in the endometrium; therefore, the low levels of IGFBP-3 in ULF during the early part of the oestrous cycle is possibly due to elevated levels of plasma oestradiol around oestrus.

The proteolysis of insulin-like growth factor binding proteins in ovine uterine luminal fluid.

During days 12-15 after oestrus (day 0), the uterine luminal fluid (ULF) of both pregnant and non-pregnant ewes contains only two prominent insulin-like growth factor binding proteins (IGFBPs) of 16-18 kDa and 22-24 kDa which preferentially bind IGF-2. Immunoblotting with an IGFBP-3 antibody revealed these to be proteolytic fragments of IGFBP-3. In contrast, the ULF from anoestrus and ovariectomized ewes contained intact IGFBP-3 (40-44 kDa) and IGFBP-2 (34 kDa). Co-incubation of ULF from an anoestrus ewe with that from a day 12 cycling ewe cleaved the IGFBP-3 present into the two lower molecular weight IGFBPs characteristic of ewes in the late luteal phase of the oestrous cycle. The variation in proteolytic activity both during the year and during the cycle suggested an influence of progesterone. Supplementation of progesterone to long-term ovariectomized ewes via a CIDR-G breeding device for 5, 10 or 15 days induced marked proteolytic activity in all 10-day treated sheep. The ULF from the 15-day treated ewes showed reduced activity and could inhibit the activity present in 10-day ULF, suggesting the induction of an inhibitor after prolonged exposure to progesterone treatment. A possible role of IGFBP-3 proteolysis in the ovine ULF may be to selectively increase the bioavailability of IGF-1 in the uterine microenvironment, which may be crucial for the rapid elongation of trophoblast that begins during days 12-15 after mating.