Illuminating the "Black Box" of Progesterone-Dependent Embryo Implantation Using Engineered Mice

Abstract

Synchrony between progesterone-driven endometrial receptivity and the arrival of a euploid blastocyst is essential for embryo implantation, a prerequisite event in the establishment of a successful pregnancy. Advancement of embryo implantation within the uterus also requires stromal fibroblasts of the endometrium to transform into epithelioid decidual cells, a progesterone-dependent cellular transformation process termed decidualization. Although progesterone is indispensable for these cellular processes, the molecular underpinnings are not fully understood. Because human studies are restricted, much of our fundamental understanding of progesterone signaling in endometrial periimplantation biology comes from in vitro and in vivo experimental systems. In this review, we focus on the tremendous progress attained with the use of engineered mouse models together with high throughput genome-scale analysis in disclosing key signals, pathways and networks that are required for normal endometrial responses to progesterone during the periimplantation period. Many molecular mediators and modifiers of the progesterone response are implicated in cross talk signaling between epithelial and stromal cells of the endometrium, an intercellular communication system that is critical for the ordered spatiotemporal control of embryo invasion within the maternal compartment. Accordingly, derailment of these signaling systems is causally linked with infertility, early embryo miscarriage and gestational complications that symptomatically manifest later in pregnancy. Such aberrant progesterone molecular responses also contribute to endometrial pathologies such as endometriosis, endometrial hyperplasia and cancer. Therefore, our review makes the case that further identification and functional analysis of key molecular mediators and modifiers of the endometrial response to progesterone will not only provide much-needed molecular insight into the early endometrial cellular changes that promote pregnancy establishment but lend credible hope for the development of more effective mechanism-based molecular diagnostics and precision therapies in the clinical management of female infertility, subfertility and a subset of gynecological morbidities.

Keywords: decidualization; endometrium; human; implantation; mediators and modifiers; mouse; progesterone; receptivity.

FIGURE 1

Cellular changes in the murine uterus during the periimplantation period. (A) On the evening of GD 4, the late-stage blastocyst attaches to the luminal epithelium of the implantation chamber (also known as the crypt) of the receptive endometrium. Note: GD 1 is defined here as the day an early morning vaginal plug is detected following overnight housing of the female with the male. The acronyms: LE, GE, AM denote luminal epithelium, glandular epithelium and antimesometrial pole respectively. (B) Following embryo attachment late on GD 4, subepithelial stromal cells surrounding the nidating blastocyst undergo extensive proliferation by the morning of GD 5. On the afternoon of GD 5, proliferating stromal fibroblasts surrounding the blastocyst differentiate to form an avascular primary decidual zone (PDZ), which initially expands toward the antimesometrial (AM) pole (see inset). At this time, the LE starts to degenerate (see inset); TE denotes the mural trophectoderm of the blastocyst, which soon breaches the LE divide to invade the underlying stroma. (C) By GD 6, the PDZ is well established, the implantation chamber epithelium is removed, and the formation of the secondary decidual zone (SDZ) surrounding the PDZ has occurred. At this time, cell proliferation is significantly decreased in the PDZ but continues in the SDZ, which expands and spreads to form the antimesometrial decidua toward the AM pole and subsequently the mesometrial decidua toward the M pole (the presumptive site for placentation). Containing terminally differentiated decidual cells, many of which are polyploid with large mono- or binuclei, the SDZ reaches full development by GD 8. In addition to decidual cells, the SDZ also includes an increased number of small blood vessels as well as a range of immune cell types (i.e., large granular uNK, macrophage and dendritic cells) that are critical for early pregnancy establishment. By GD 8, the PDZ is markedly degenerated, and from this day onwards, placental and embryonic expansion progressively replaces the SDZ. Such an expansion transforms the antimesometrial decidua to a thin layer of cells termed the decidua capsularis. With pregnancy progression, the mesometrial decidua thins to the decidua basalis to accommodate the enlarging placenta, containing the embryonic-derived spongiotrophoblast cell layer and labyrinth zone. Blood vessels are denoted by BV. The dotted arrow indicates the direction of SDZ expansion. Elements of this artwork were adapted in modified form with permission from Lim and Wang (2010).

FIGURE 2

Signaling cross talk drives progesterone-dependent endometrial receptivity and decidualization. (A) Spanning the epithelial-stromal divide of the endometrium, the progesterone-PGR-IHH-COUP-TFII signaling axis primarily controls ESR1 activity in the epithelium. Suppression of ESR1 activity is a prerequisite for luminal epithelial differentiation of the endometrium to be receptive to embryo attachment and implantation. Note: Msx 1 and 2—and many other important mediator signals—that are important in this process are not shown. (B) A representative signaling network that is essential for progesterone driven endometrial stromal cell decidualization is shown. The dotted arrows indicate signaling relationships for which direct regulatory control has yet to be established. Adapted with permission in modified form from Wu et al. (2018).

FIGURE 3

The GATA2 transcription factor is a direct epithelial modifier of PGR-mediated transcription in the murine endometrium. (A) Quantitative real-time PCR analysis reveals that absence of GATA2 in the endometrium of the ovariectomized mouse results in a significant reduction in Pgr transcript levels. Note: in the uterus of an ovariectomized control mouse, the majority of Pgr expression in located in the epithelial compartment (Rubel et al., 2016). Also note that the GATA2 transcription factor is conditionally ablated in PGR positive cells of the uterus in the Gata2^d/d mouse. In silico analysis highlights numerous candidate DNA binding motifs for GATA2, which are located throughout the 5′ regulatory regions of the murine Pgr gene (red boxes) and red highlighted sequences. Chromatin immunoprecipitation and in vitro transient transfection experiments confirmed that many of these binding sites directly bind GATA2 and are functional (Rubel et al., 2016). (B) Using uterine tissue from ovariectomized mice treated with progesterone for 6 h, chromatin immunoprecipitation followed by genome-wide sequencing (ChIP-seq) identified enriched binding motifs for the GATA2 and PGR transcription factors within GATA2 binding intervals throughout the genome. The Venn diagram displays the progesterone responsive genes in the mouse uterus that contain binding sites for PGR or GATA2 within ± 25 kb of gene boundaries. Note: the significant overlap (935 genes) that represent genes jointly bound by PGR and GATA2 (Rubel et al., 2016). (C) The traces show the co-occupancy locations of the PGR, GATA 2, SOX17, and FOXA2 transcription factors at a distal enhancer region on the murine Ihh gene (Wang et al., 2018). Note: SOX17 is also a direct target of GATA2 and PGR (Wang et al., 2018); the FOXA2 transcription factor (not covered in this review) is critical for uterine glandular epithelial function that is required for pregnancy establishment (Kelleher et al., 2017). With permission, parts of this schematic were reproduced in modified form from Rubel et al. (2016) and Wang et al. (2018). **p < 0.01.

FIGURE 4

The SRC-2 coactivator is required for progesterone-dependent acceleration of the glycolytic flux that drives endometrial stromal cell decidualization. (A) The pleiotropic properties of SRC-2 are based on its complex protein functional domain organization. Activation domains 1–3 (AD1-3), receptor interaction domain (RID), basic helix-loop-helix domain (bHLH), the Per/ARNT/Sim domains -A and -B (PAS-A and –B), leucine-X-X-leucine-leucine [X denotes any amino acid (LXXLL)], and the glutamine-rich/interaction motif (Q-rich/IM) are indicated. (B) To generate sufficient numbers of epithelioid decidual cells to support embryo implantation, stromal fibroblasts of the endometrium rapidly proliferate in response to progesterone prior to their differentiation. By markedly increasing the rate of glucose uptake and glycolysis, an endometrial stromal cell rapidly produces two daughter cells following mitosis. Glycolysis from glucose to pyruvate is referred to as the glycolytic flux. Increasing the glycolytic flux serves to rapidly provide the necessary bioenergy and biomolecules to meet the urgent demands of a proliferating endometrial stromal cell that is about to form two daughter cells. The progesterone-dependent acceleration of the glycolytic flux requires SRC-2 co-regulation of PGR-mediated induction of PFKFB3 (Kommagani et al., 2013), a potent positive regulator of the glycolytic flux. Using its kinase domain, PFKFB3 converts fructose-6-P to fructose-2, 6-P, which allosterically activates PFK-1, a pivotal checkpoint of glycolysis. Increasing the glycolytic flux results in a net gain of two ATP molecules per glucose molecule catabolized, and the generation of glycolytic intermediates (i.e., glucose-6-P and pyruvate) to furnish the required precursors for macromolecular and organelle biosynthesis by downstream anabolic pathways. 6-Phosphofructo-2-kinase/fructose-2, 6-bisphosphatase 3, phosphofructokinase-1 are abbreviated by PFKFB3 and PF-1, respectively. With permission, aspects of this figure were reproduced in modified form from Szwarc et al. (2014b).