Asynchronous Embryo Transfer Followed by Comparative Transcriptomic Analysis of Conceptus Membranes and Endometrium Identifies Processes Important to the Establishment of Equine Pregnancy

Abstract

Preimplantation horse conceptuses require nutrients and signals from histotroph, the composition of which is regulated by luteal progesterone and conceptus-secreted factors. To distinguish progesterone and conceptus effects we shortened the period of endometrial progesterone-priming by asynchronous embryo transfer. Day 8 embryos were transferred to synchronous (day 8) or asynchronous (day 3) recipients, and RNA sequencing was performed on endometrium and conceptuses recovered 6 and 11 days later (embryo days 14 and 19). Asynchrony resulted in many more differentially expressed genes (DEGs) in conceptus membranes (3473) than endometrium (715). Gene ontology analysis identified upregulation in biological processes related to organogenesis and preventing apoptosis in synchronous conceptuses on day 14, and in cell adhesion and migration on day 19. Asynchrony also resulted in large numbers of DEGs related to 'extracellular exosome'. In endometrium, genes involved in immunity, the inflammatory response, and apoptosis regulation were upregulated during synchronous pregnancy and, again, many genes related to extracellular exosome were differentially expressed. Interestingly, only 14 genes were differentially expressed in endometrium recovered 6 days after synchronous versus 11 days after asynchronous transfer (day 14 recipient in both). Among these, KNG1 and IGFBP3 were consistently upregulated in synchronous endometrium. Furthermore bradykinin, an active peptide cleaved from KNG1, stimulated prostaglandin release by cultured trophectoderm cells. The horse conceptus thus responds to a negatively asynchronous uterus by extensively adjusting its transcriptome, whereas the endometrial transcriptome is modified only subtly by a more advanced conceptus.

Keywords: asynchronous embryo transfer; conceptus; endometrium; exosome; horse; transcriptome.

Figure 1

Hierarchical cluster analysis of differentially expressed genes (DEGs) with a log2 fold change higher than 1.4 or lower than -1.4 for equine endometrial samples collected after transfer of day 8 horse embryos to synchronous (day 8: SY) or asynchronous (day 3: ASY) recipient mares. Endometrial samples were collected on day 14 (d14) or 19 (d19) of embryo development (blue = downregulated genes; red = upregulated gene). The sample number is indicated by a number from 1 to 16.

Figure 2

Venn diagram for the overlap of DEGs in the endometrium among three pair-wise comparisons, day 14 of conceptus development between synchronous (day 14 endometrium) and asynchronous (day 9 endometrium; D14 syn/asyn), on day 19 of conceptus development between synchronous (day 19 endometrium) and asynchronous (day 14 endometrium; D19-syn/asyn) and between day 14 synchronous and day 19 asynchronous pregnancies (D14-sync/D19-asyn). The numbers of DEGs in each compartment is depicted.

Figure 3

Pair-wise correlation heat map (A) and hierarchical cluster (B) analysis of DEGs with a log2 fold change higher than 1.4 or lower than -1.4 for equine conceptus membrane samples recovered after transfer of day 8 embryos to a synchronous (day 8: SY) or asynchronous (day 3: ASY) recipient mare, and collected on day 14 (d14) or 19 (d19) of conceptus development (blue = downregulated genes; red = upregulated genes). The sample number is indicated by a number from 1 to 16.

Figure 4

Venn diagram for the overlap of DEGs in equine conceptus membranes between three pair-wise comparisons, day 14 of pregnancy between synchronous (day 14 recipient) and asynchronous (day 9 recipient; D14 syn/asyn) pregnancies, on day 19 of pregnancy between synchronous (day 19) and asynchronous (day 14 recipient; D19-syn/asyn) and between day 14 synchronous day 19 and asynchronous (D14-sync/D19-asyn). The numbers of DEGs in each compartment is depicted.

Figure 5

Gene ontology analysis for genes upregulated and downregulated in equine endometrium on day 14 and 19 of pregnancy between synchronous (recipient and donor ovulated on same day) and asynchronous (recipient ovulated 5 days after the donor) embryo transfer (ET; D14 syn/asyn and D19 syn/asyn). The major categories within the biological process, cellular component and molecular function are represented by the number of genes in the category, selected on the basis of a p-value < 0.05.

Figure 6

Gene ontology analysis for genes upregulated and downregulated in equine conceptus membrane, on day 14 and 19 of pregnancy between synchronous (recipient and donor ovulated on same day) and asynchronous (recipient ovulated 5 days after the donor) ET (D14 Syn/Asyn and D19 Syn/Asyn). The major categories within the biological process, cellular component and molecular function are represented by the number of genes involved in the category, selected on the basis of a p-value < 0.05.

Figure 7

Effect of IGFBP3 on attachment (A) and proliferation (B) of cultured equine trophectoderm cells. For the attachment assay, wells were coated with BSA (negative control: 1000 ng/mL), fibronectin (positive control) or IGFBP3 (62, 125, 250, 500 and 1000 ng/mL). After 2 h of culture, the number of cells attached was counted and the average number of cells attached per mm² (± s.e.m) calculated. The same letter indicates no significant difference (p > 0.05) in cell attachment in the presence of fibronectin (a,b,c) or IGFBP3 (x,y). Both fibronectin and IGFBP3 stimulated a dose-dependent increase in cell attachment, although the effect of the highest concentration of fibronectin was much larger. For the proliferation assay, cells were cultured for 30 h and then treated with IGFBP3 (1, 10, 100 and 1000 ng/mL), normal culture media (positive control), and serum-free media (negative control) for 24 h. BrdU was used to identify proliferating cells. The total number of cells (DAPI), and proliferating cells (BrdU positive) were counted, and the percentage of proliferating cells (± s.e.m) per mm² was calculated. The analysis was performed for three independent experiments, with three replicates per condition; 5 non-overlapping sections were imaged per condition. Cell proliferation was higher in the presence of serum, but was not further increased by the addition of IGFBP3.

Figure 8

Effect of bradykinin on prostaglandin synthesis by equine trophectoderm cells. Cells were incubated in the presence of bradykinin (1, 10, 100 and 1000 ng/mL) for 6 and 24 h or with serum free medium for 6 h (negative control). The concentrations of PGE2, PGF2α and 6-keto PGF1α (pg/mL) were measured in the culture media, and are depicted as the fold change for a given treatment versus the negative control. The analysis was performed on three independent experiments, with three replicates per condition. Bradykinin stimulation resulted in increased production of all prostaglandins. After 24 h, there was evidence of a bradykinin dose dependent increase for PGE2 secretion (p < 0.0005) and a tendency for a dose dependent increase in PGF2α secretion (p = 0.055), but not in 6-keto PGF1α secretion.

Figure 9

Day 8 embryos were transferred to recipient mares that ovulated on the same day (synchronous: n = 10) or 5 days after (asynchronous: n = 10) the donor mare. Conceptuses were recovered 6 or 11 days later (5 per group), when the conceptus would be at either 14 or 19 days of development and the recipient mares would be at either 9, 14 or 19 days after ovulation. Adapted from [40].