52 Equine embryo size does matter!

The size of equine embryos collected on the same day post-ovulation (dpo) is very variable. So far, gene expression related to embryo size has not been explored. The aim of this study was to evaluate gene expression in relation to embryo size in equine blastocysts. Non-nursing Saddlebred mares located on two experimental farms were inseminated with the semen of one unique stallion per farm after human chorionic gonadotrophin (hCG) ovulation induction and detection within 48 h on one farm and 24 h on the other farm. At 8 dpo (9–10 days post-hCG), 17 and 11 embryos in each farm, each from a different dam, were recovered by uterine flushing. Embryo diameter was measured and each embryo was bisected to obtain samples of pure trophectoderm (TE) or inner cell mass enriched (TE-ICM) trophoblast. RNA expression was analysed separately in both parts of nine small (S, <700, mean of 560 ± 85 µm), 11 medium (M, 700–1200, mean of 886 ± 160 µm) and eight large (L, >1200, mean of 1719 ± 488µm) embryos, using paired end, non-oriented RNA-sequencing (NextSEqn 500; Illumina). To discriminate gene expression in the ICM from that in the TE, deconvolution (DeMixT) was used on the TE-ICM data. Differential expression was analysed (DESEqn 2), with farm and embryo sex as cofactors using a false discovery rate (FDR) <0.05 cutoff. Overrepresentation test was performed on differentially expressed genes using PANTHER web software (http://www.pantherdb.org/) with the Gene Ontology BP database. Out of the 14 249 and 13 406 genes, respectively, expressed in ICM and TE, 642 and 1228 in ICM and 123 and 443 genes in TE were differentially expressed when comparing, respectively, S vs. M and L vs. M. Of particular interest, in the ICM and TE of S, insulin-like growth factor (IGF)1 (log2 fold change (log2FC) = 1.9 and 2.0 for ICM and TE, respectively, FDR < 0.05) was more expressed, whereas in L, IGF1 was up-regulated in ICM (log2FC = 2.0, FDR < 0.05) and down-regulated in TE (log2FC = −1.2, FDR < 0.05), both compared with M. IGF1 is known to stimulate oestrogen production in pig embryos. Here, several P450 cytochromes including cytochrome P450 aromatase (CYP19A1, log2FC = 4.46, FDR < 0.0001) associated with 17β-hydroxysteroid dehydrogenase (HSD17B1, log2FC = 0.58, FDR < 0.05) were up-regulated in the ICM in L compared with M. Moreover, Nanog Homeobox gene (log2FC = −3.0, FDR < 0.0001) and several SRY-Box Transcription Factors (SOX), including SOX2 (log2FC = −2.3, FDR < 0.01), were down-regulated in the ICM of L compared with M. For the ICM, differentially expressed genes in S vs. M were principally related to mitochondria and regulation of apoptosis, whereas in L vs. M, development, cell proliferation, and lipid transport were mainly represented. In TE, no gene set was overrepresented in S vs. M, whereas tube development and organitrogen compound metabolic process were overrepresented in L vs. M. In conclusion, equine embryos of different diameter also differ in developmental stage. S embryos express more genes associated with mitochondria and apoptosis regulation, whereas L embryos seem involved in cell lineage development and fast growth. The ICM of L embryos expresses genes involved in oestrogen production which, together with IGF1, may participate in the early signal for maternal recognition of pregnancy.