118 Influence of sire fertility status on conceptus-induced transcriptomic response of the bovine endometrium

The aim of this study was to examine the effect of sire fertility status on conceptus-induced changes in the endometrial transcriptome. Holstein Friesian bulls (3 high fertility, HF; three low fertility, LF) were selected from the Irish national population of AI bulls (minimum of 500 inseminations/bull) based on adjusted fertility scores (HF: +4.37% and LF: −12.7%; mean = 0%). To generate elongated conceptuses, Day 7 blastocysts produced in vitro using sperm from these six bulls were transferred in groups of 5 to 10 to synchronised heifers (n = 7 heifers per bull; total 42 heifers). Conceptuses were recovered following slaughter on Day 15 (recovery rate: HF 59.4% vs. LF 45.0%; P < 0.05). In parallel, Day 15 endometrial explants were recovered from synchronised cyclic heifers (n = 4). Explants from each heifer were co-cultured for 6 h in RPMI medium with (i) nothing, Control; (ii) 100 ng mL⁻¹ ovine recombinant interferon tau (IFNT); (iii) a single conceptus from each HF bull; or (iv) a single conceptus from each LF bull. To minimise variation, explants from the same uterus were used across all treatments, replicated across four heifers. After 6 h, explants were snap frozen and stored at −80°C. Extracted mRNA was subjected to RNA-seq (Illumina NextSEqn 500) and the resulting data were analysed through a bioinformatic pipeline with R software (https://www.r-project.org/). An initial unsupervised analysis highlighted the effect of individual heifer (explant donor) on the endometrial transcriptome. Thus, data were adjusted with the Surrogate Variable Analysis package before determining differentially expressed genes (DEG) with the edgeR package. The numbers of DEG (fales discovery rate (FDR) < 0.05) were HF vs. Control: 956; LF vs. Control: 1021; IFNT vs. Control: 1301; HF vs. LF: 2. Unsurprisingly, the majority of DEG (658) were common to all comparisons and were related to IFNT production. Before applying the adjusted P-value, there were 700 DEG between HF and LF, with 191 and 509 genes more expressed in HF or LF, respectively (P < 0.05). Overrepresentation analysis of Kyotoo Encyclopedia of Genes and Genomes (KEGG) pathways (FDR < 0.05; DAVID software) revealed that DEG with higher expression in LF were involved only in cell cycle and proteolysis. However, up-regulated DEG by HF conceptuses were strongly associated with immune process pathways, such as tumor necrosis factor, nuclear factor kappa B, cytokine-cytokine receptor interaction, and toll-like receptor signalling. These pathways were also enriched by up-regulated DEG by IFNT treatment compared to control. Furthermore, only the HF group, and not the LF group, affected the expression of most genes on these pathways (P < 0.05) according to a negative binomial regression model. Finally, a weighted gene co-expression network analysis (WGCNA package) revealed two clusters of co-expressed genes associated with the HF conceptuses (P < 0.05), which were also enriched for the above-mentioned pathways. In conclusion, while the differences induced in the endometrial transcriptome by conceptuses from HF and LF bulls were subtle, HF conceptuses, similar to IFNT treatment, stimulated pathways involved in immune response, which were apparently not dysregulated by LF conceptuses.

This research was funded by Science Foundation Ireland (16/IA/4474).