A burst of genomic innovation at the origin of placental mammals mediated embryo implantation

Abstract

The origin of embryo implantation in mammals ~148 million years ago was a dramatic shift in reproductive strategy, yet the molecular changes that established mammal implantation are largely unknown. Although progesterone receptor signalling predates the origin of mammals and is highly conserved in, and critical for, successful mammal pregnancy, it alone cannot explain the origin and subsequent diversity of implantation strategies throughout the placental mammal radiation. MiRNAs are known to be flexible and dynamic regulators with a well-established role in the pathophysiology of mammal placenta. We propose that a dynamic core microRNA (miRNA) network originated early in placental mammal evolution, responds to conserved mammal pregnancy cues (e.g. progesterone), and facilitates species-specific responses. Here we identify 13 miRNA gene families that arose at the origin of placental mammals and were subsequently retained in all descendent lineages. The expression of these miRNAs in response to early pregnancy molecules is regulated in a species-specific manner in endometrial epithelia of species with extreme implantation strategies (i.e. bovine and human). Furthermore, this set of miRNAs preferentially target proteins under positive selective pressure on the ancestral eutherian lineage. Discovery of this core embryo implantation toolkit and specifically adapted proteins helps explain the origin and evolution of implantation in mammals.

Fig. 1. Regulation of core miRNA toolkit by early pregnancy markers.

The endometrial epithelium of bovine (superficial implantation strategy) (top panel) or human (invasive implantation strategy) (bottom panel) are regulated by molecules important for endometrial function in early pregnancy in eutheria (P4: progesterone; hCG: human chorionic gonadotrophin; Interferon Tau: IFNT; Macrophage capping protein: CAPG, and protein disulfide isomerase: PDI).

Fig. 2. Phylogenetic distribution of the miRNA families and their levels of targeting in positively selected genes.

a Phylogenetic distribution of miRNA families specific to therian and eutherian mammals. The mammal phylogeny displays the species sampled in our analysis. The corresponding matrix shows the presence (dark red) or absence (pale grey) of miRNA families across the species sampled. b Violin plot comparing the number of target miRNA binding sites from the core miRNA toolkit per transcript in the 84 genes that underwent positive selection on the stem Eutherian lineage (PSGs) (left) compared sampled sets of non-positively selected but randomly sampled genes that are targeted by the core toolkit of miRNAs (nPSG) (right). For each of the 84 PSGs, the mean number of predicted miRNA binding sites was determined for each target transcript in humans (depicted in green). This was compared to the mean number of binding sites for each of the nPSGs (depicted in blue). The mean number of binding sites was determined to be significantly different between the two datasets when p ≤ 0.05, two-sample t-test with unequal variance.

Fig. 3. Regulation of stem lineage miRNAs in human and bovine endometrial epithelial cells treated with hCAPG or bCAPG respectively.

Expression of stem lineage miRNAs miR-432-5p, miR331-3p, miR-671-5p, miR-671-3p, miR-340-5p, miR-188-5p, miR432-3p and miR-324-5p in either human (left-hand side of each pair) or bovine (right-hand side of each pair) endometrial epithelial cells (n = 6 per treatment). Primary bovine endometrial epithelial cells were treated with vehicle control (grey circle), or 1000 ng/μL bCAPG (purple circle) for 24 h. Human Ishikawa immortalised endometrial epithelial cells were treated with control (open circle), vehicle control (closed circle), or 1000 ng/μL hCAPG (purple circle) for 24 h. Significant differences in miRNA expression values determined when p ≤ 0.05 are depicted by an asterisk (*).

Fig. 4. Regulation of stem lineage miRNAs in human and bovine endometrial epithelial cells treated with hPDI or bPDI, respectively.

Expression of stem lineage miRNAs mir-324-5p, miR-542-3p, miR-185-5p, and miR151a-3p in either human (left-hand side of each pair) or bovine (right-hand side of each pair) endometrial epithelial cells (n = 6 per treatment). Primary bovine endometrial epithelial cells were treated with vehicle control (grey circle), or 1000 ng/μL bPDI (orange circle) for 24 h. Human Ishikawa immortalised endometrial epithelial cells were treated with control (open circle), vehicle control (closed circle), or 1000 ng/μL hPDI (orange circle) for 24 h. Significant differences in miRNA expression values determined when p ≤ 0.05 are depicted by an asterisk (*).