Maternal age affects equine Day 8 embryo gene expression both in trophoblast and inner cell mass

Abstract

Background: Increased embryo loss as mares become older is a major consideration for breeders as older animals are currently used for reproduction in the equine industry. Lower embryo quality has been pointed out as partly responsible for this reduced fertility. Here, the effect of mare's age on blastocysts' gene expression was explored. Day 8 post ovulation embryos were collected by uterine flushing from multiparous young (YM, 6-year-old, N = 5) and older (OM, > 10-year-old, N = 6) non-nursing Saddlebred mares, inseminated with the semen of one stallion. Pure (TE_part) or inner cell mass enriched (ICMandTE) trophoblast were obtained by embryo bisection and paired end, non-oriented RNA sequencing (Illumina, NextSeq500) was performed on each hemi-embryo. To discriminate gene expression in the ICM from that in the TE, deconvolution (DeMixT R package) was used on the ICMandTE dataset. Differential expression was analyzed (DESeq2) with embryo sex and diameter as cofactors using a false discovery rate < 0.05 cutoff. Functional annotation and classification of differentially expressed genes and gene set enrichment analysis were also performed. Results: Maternal aging did not affect embryo recovery rate, embryo diameter nor total RNA quantity but modified gene expression in equine D8 embryos. The expression of genes in the ICM seemed to be more altered by maternal aging than in the TE. In both compartments, the expression of genes involved in mitochondrial function, translation and transcription due to chromatin modification were disturbed by maternal age. Mitosis, signaling and adhesion pathways and embryo development were particularly decreased in the ICM hemi-embryos from old mares. Finally, in TE, ion movement related genes were affected. Conclusions: This is the first study showing an effect of maternal age on gene expression in the equine blastocyst at Day 8 post ovulation. Maternal age, even for mares as young as 10 years old, disturbs mitosis, translation and transcription, cell signaling and adhesion as well as mitochondrial function and cell commitment of horse embryos. These perturbations may affect further embryo development and contribute to decreased fertility due to aging.

Maternal parity affects Day 8 embryo gene expression in old mares

“…Embryo were sexed using X Inactive Specific Transcript (XIST) expression as previously described [23].…”

“…Oocytes are particularly affected by maternal age with alterations of spindle stability [5][6][7], altered gene expression [8,9], and altered metabolism [8,[10][11][12] being reported, all suggesting that oocyte developmental potential is reduced in old mares. The resulting embryos were smaller at the same developmental age in most studies [13][14][15][16][17][18][19][20][21][22][23] with altered gene expression [23] and metabolism [11,12], suggesting impaired development that has mainly been related to the oocyte quality. The reproductive tract, however, is also affected by maternal age.…”

“…As previously described [23], alignment was performed using STAR version 2.6 [44] on previously modified Ensembl 99 EquCab3.0 assembly and annotation. Genes were then counted with FeatureCounts [45] from Subreads package version 1.6.1.…”

“…The deconvolution method has already been described in equine embryos [23]. Briefly, this method enables the estimation of the relative gene expression of TE and ICM cell types within the hemi-embryo ICMandTE which is composed of both trophoblast and inner cell mass in unknown relative proportions.…”

“…TE_part were compared with Deseq2 version 1.28.1 [52] to confirm that the deconvolution was effective at separating gene expression. To check if deconvolution was efficient, as previously described [23], the expression of several genes proper to ICM and TE cells in equine embryos identified using literature search [53] was compared before and after deconvolution. Results of these analyses were represented through manually drawn Venn diagrams as well as principal component analysis graphics of individuals, using ggplot2 version 3.3.3 [54] and factoextra version 1.0.7 [55].…”