Endocrine Journal
Online ISSN : 1348-4540
Print ISSN : 0918-8959
ISSN-L : 0918-8959
REVIEW
Epithelial mesenchymal transition in human menstruation and implantation
Hiroshi Uchida
Author information
JOURNAL OPEN ACCESS FULL-TEXT HTML

2024 Volume 71 Issue 8 Pages 745-751

Details
Abstract

The endometrium during the sexual cycle undergoes detachment, tissue remodeling, and differentiation during the menstrual cycle. Localized and transient destruction and regeneration of endometrial tissue are also essential for pregnancy. It is possible to attribute many causes of failure in infertility treatment to the implantation stage. To improve the success rate of plateau fertility treatment, it is important to understand the regeneration mechanism of the endometrium, a unique regenerative tissue in the human body. In association with cell proliferation, tissue remodeling requires the relocation of proliferative cells, and the steady-state epithelial cells need to be motile for the relocation. Transient add-on motile activity in epithelial cells is mediated by epithelial to mesenchymal transition (EMT) and reversible mesenchymal to epithelial transition (MET). The destruction and regeneration of endometrial tissue over a period of days to weeks requires a system with a rapid and characteristic mechanism similar to that of wound healing. Here, I review the relationship between the well-known phenomenon of EMT in wound healing and endometrial tissue remodeling during the sexual cycle and pregnancy establishment, which are automatically triggered by menstruation and embryonal invasion.

Introduction

In humans, the successful pregnancy of a healthy couple occurs in approximately 20% of cases per period. In all conceptions, around 30% of trials fail at the stage of implantation, another 30% are lost in the early stage of pregnancy, and approximately 15% result in early miscarriage. Through in vitro fertilization and embryo transfer, the live birth rate still reaches only 25%–30%. The in vitro fertilization rate of gametes reaches around 80% by improving laboratory techniques; however, the successful rate of embryo transfer does not increase much after reaching a plateau of 30%–40%. The major reason for loss is implantation failure. Limited information has been obtained about the mechanisms of human implantation because the field of implantation is located within the cavity of the uterus, which is physiologically and ethically difficult to approach in experiments. Although pregnancy is established in such a space, it is useful for infertile couples to unveil the complex mechanism of implantation. Endometrial cell dynamics during implantation should be elucidated to improve live birth rates.

After menarche, the endometrial tissue monthly repeats destruction and regeneration over four hundred times during more than thirty years. Unlike local metabolism, endometrial shedding and regeneration is a massive and frequent tissue renewal mechanism designed to be cyclic, and it is quite specific and characteristic of the female body. The tissue remodeling process is similar to early development at the common point of appropriate differentiation and repositioning of cells. Furthermore, the set of destruction and repair is observed during the wound healing process. Given that epithelial mesenchymal transition (EMT) is known to be a common mechanism for early development [1] and wound healing processes [2], it is expected that EMT (and mesenchymal epithelial transition (MET) as a reversible mechanism) is also used in endometrial remodeling. In fact, many studies have shown that EMT is involved in the remodeling of the endometrium during menstruation and the establishment of pregnancy.

EMT/MET

Human tissue is composed of epithelial cells and stroma transformed through epithelial mesenchymal transition (EMT): a phenomenon in which epithelial cells acquire mesenchymal cell traits. It is observed during early embryonic development [1], organogenesis [3], tissue repair [2], and carcinogenesis [4]. The reverse process is known as mesenchymal epithelial transition (MET). Epithelial cells are characterized by tight contact with neighboring cells and apicobasal polarity through the cell-cell contact of adherence junctions, desmosomes, and tight junctions. Therefore, protein markers of epithelial cells include proteins consisiting of cell-cell junctions, such as E-cadherin (adherence junctions), cytokeratin (desmosomes), and occludin and claudin (tight junctions) (Fig. 1). Tight cell-cell contact results in a plastic and immotile state in epithelial monolayers. In contrast, mesenchymal cells are multipolar, spindle-shaped, and motile because cell-cell contact is loose with the connective tissue or extracellular matrix. Loose cell-cell contact protein (N-cadherin), intermediate filament protein (vimentin), connective tissue protein (fibronectin), and transcription factor proteins causing EMT (Zeb1 and Snail) are known as protein markers of mesenchymal cells [5].

Fig. 1

Characteristics of EMT

Epithelial cells are plastic and suitable for maintaining tissue shape, whereas mesenchymal cells have dynamic characteristics that are advantageous for flexible modification of tissue structure. The key to these characteristics is cell motility, which is characterized by cell-cell adhesion patterns and the presence or absence of apico-basolateral cell polarity. Cell shape is defined by the cell-cell adhesion pattern and cell polarity, with epithelial cells being polygonal and mesenchymal cells being unsettled. EMT is a phenomenon that enhances cell motility by loosening cell-cell adhesion, resulting in a change from static epithelial cells with E-cadherin-based cell adhesion mechanisms to dynamic mesenchymal cells with relatively loose N-cadherin-based cell adhesion mechanisms. This characteristic change is observed at the place of embryonic penetration, and the degree of EMT is particularly high and gradually so in epithelial cells near the embryo.

EMT in endometrial cycle

The human endometrium is composed mainly of glandular epithelial cells and stromal cells, along with a diverse group of other cells, including vascular endothelial, vascular smooth muscle, haematopotic, and immune cells. Histologically, it is composed of glandular epithelium, luminal epithelium, and stroma, with a basal layer on the luminal side of the myometrium and a laminated functional layer (Fig. 2A). In the case of the endometrium, the functional layer is detached and shed during the menstrual cycle, leaving the basal layer. Regeneration of the functional layer occurs via tissue stem cells in the remaining basal layer or in the functional layer that has escaped detachment. Besides menstruation, tissue remodeling and regeneration of the endometrium occurs also in inflammatory deconstruction, injury, and idiopathic insult.

Fig. 2

EMT in menstruation and implantation

A) The endometrium, which coats the uterine cavity, under endocrinologic control, undergoes a renewal cycle of detachment and regeneration during the sexual cycle. The endometrium is identified as a basal layer and a functional layer, with menstruation resulting in the detachment of most of the functional layer. After menstruation, the disrupted endometrium is regenerated during the proliferative phase by cell proliferation and tissue differentiation derived from tissue stem cells. After ovulation, the endometrium enters the secretory phase under the influence of progesterone and undergoes increased differentiation, transforming into endometrial tissue suitable for pregnancy. The specific differentiation of endometrial stromal cells during this period is called decidualization, and the most suitable phase for pregnancy is called the window of implantation (WOI).

B) Human implantation can be divided into four sequential processes: embryonal apposition, adhesion, penetration, and invasion. These four steps are also defined as apposition, adhesion, destruction and reconstruction for the endometrial epithelial cell barrier. After ovulation, prior to the apposition stage, EMT is already initiated in endometrial epithelial cells, which lose their characteristic apico-basolateral polarity and undergo shape changes such as flattening to gain receptivity under the control of ovarian steroid hormones, such as estradiol and progesterone. EMT accelerated by close cell adhesion to the embryo is also observed as cell adhesion molecules, which were previously E-cadherin-dominant, change their expression to N-cadherin-dominant, and endometrial epithelial cells become more motile. The characteristic motility is a movement distal to the embryonic center, which is observed as an embryonic evasive maneuver, as if to form a route of embryonic penetration. The passage of the embryo results in the loss of signals via adhesion molecules that connect to the embryo, or the decrease of paracrine factors due to distance separation. Finally, remodeling of endometrial epithelial cells is completed by tissue remodeling, through reversible MET provoked by loss of EMT-inducing signals.

EECs, endometrial epithelial cells; ESCs, endometrial stromal cells

Transition between epithelial and mesenchymal cells is an important mechanism in endometrial remodeling and regeneration after menstruation, and in endometrial stromal differentiation, (i.e., decidualization), enabling the conditions for a successful pregnancy. When analyzing endometrial cells using primary culture, it is common to use cytokeratin-positive cells for endometrial epithelial cells and vimentin-positive cells for endometrial stromal cells, using the marker proteins described above. Even after screening with high purity, the purity of the cell material gradually decreased through repeated passages. This event also indicates that endometrial cells are constantly undergoing repeated EMT/MET.

The menstrual cycle in the endometrium can be divided into two phases (Fig. 2A). In the proliferative phase, the ovarian steroid hormone estradiol proliferates endometrial cells, followed by menstruation. After ovulation, progesterone-derived decidualization of endometrial stromal cells occurs in preparation for receiving fertilized eggs in the secretory phase. Without pregnancy, the reconstructed endometrium is broken and shed again during menstruation. In addition to ovarian steroid hormones, numerous factors, including hormones, cytokines, and miRNAs are coordinately regulated in this repeated regeneration system.

During this remodeling process, it is thought that EMT/MET plays a key role. Interestingly, it has been reported that N-cadherin-positive progenitor cells, which also express E-cadherin and vimentin (epithelial cell marker proteins), localize in glandular epithelial cells [6]. The mechanism of endometrial regeneration remains unclear, however, remodeling of endometrial epithelial and stromal tissue is thought to originate from endometrial stromal cells. Several reports indicate that stream progenitor cells localize to both endometrial epithelial and stromal tissues [7, 8]. Furthermore, after progesterone withdrawal, it is demonstrated that the up-regulation of several EMT marker proteins is associated with the decreased expression of MET marker proteins [9].

It is demonstrated that decidualization is partly controlled by MET [10]. Under the stimulation of ovarian steroid hormone progesterone, the human endometrial stromal cell is morphologically changed from a spindle shape to an epithelial-like rounded and flattened shape, and in in vitro culture, the change in cell shape is reversed by withdrawal of ovarian steroid hormones and cAMP [11]. Decidualized morphological change in endometrial stromal cells and increased secretion of prolactin and insulin-like growth hormone binding protein 1 (IGFBP-1), which are differentiation marker proteins of decidualized stromal cells, can be reproduced by histone deacetylase inhibitors (HDACIs), such as trichostatin A without progesterone exposure [12]. Maintaining healthy endometrial tissue via EMT/MET control is important for receptivity. If the appropriate mechanism does not function, the endometrial tissue changes to fibrotic, causing a condition known as Asherman’s syndrome [13].

EMT in implantation

Numerous signals caused by hormones, cytokines, growth factors, and miRNAs coordinately control a series of implantation processes in humans. To ensure a successful pregnancy, the embryo must pass the barrier of endometrium, consisting of endometrial epithelial cells. There must be reasonable cell dynamics, involving morphological change and cell motion in both embryonal trophectoderm cells and maternal endometrial cells, therefore, representative EMT is provoked in both.

During human implantation, embryos reach into endometrial stromal spaces through the endometrial epithelial cell barrier via several steps of cell dynamics, including apposition, adhesion, penetration, and invasion. At the same time, the endometrial epithelial cell sheets process the sequential steps of apposition, adhesion, destruction, and reconstruction one by one (Fig. 2B). Not limited to the endometrium, the epithelial cell layer forms a non-adherent physiological and immunological barrier to the luminal surface, while they form a hierarchical structure to the inner stromal cell tissue, via the basement membrane, and the epithelial cell layer on the contralateral luminal surface maintains the luminal structure by not adhering to each other at the apical side of them. The difference in the relationship with the adjacent environment is defined by the polarity of the epithelial cells. On the apical side, there are channels for material exchange, various receptors for signal transduction, non-adherent molecules, and cilia for immunological defense. On the lateral side, adhesion mechanisms, such as tight junctions, adherence junctions, and desmosomes, are formed to control the exchange of substances between neighboring cells or from the lumen into the tissue to maintain cell shape by connecting with the cytoskeleton. In contrast to epithelial cells, mesenchymal cells are nonpolar, loosely connect to neighboring cells and the extracellular matrix, and keep high motile state. During implantation, endometrial epithelial cells are required to acquire receptivity [14]; in other words, they are required to acquire the ability to adhere to the embryonic epithelial cells at the apical side and to create a route for passage of the adherent embryo to the stromal tissue. EMT, which allows for the simultaneous relinquishment of cell polarity and acquisition of motility, appears to be extremely well suited to achieve both of these requirements.

Prior to the apposition phase, endometrial epithelial cells prepare to accept the embryo. In the mid-secretory phase, progesterone differentiates endometrial stromal cells (decidualization), as described above. Also in endometrial epithelial cells, ovarian steroid hormones, estradiol and progesterone can induce differentiation to gain receptivity. Wide-spread and flattened morphological change and increased expression of glycogen can be observed in endometrial epithelial cells by treating with both of estradiol and progesterone in vitro [15]. Down-regulation of E-cadherin (epithelial cell marker protein) and up-regulation of N-cadherin (mesenchymal cell marker protein) is synchronized in EMT cells, and the characteristic alteration of protein expression is called “cadherin switch.” Without an embryo, cadherin switch is weakly observed in endometrial epithelial cells treated by estradiol and progesterone [16]. It is thought that ovarian steroid hormone triggers the initiation of EMT in endometrial epithelial cells. Interestingly, the morphological change and cadherin switch in endometrial epithelial cells is significantly enhanced by embryo attachment and adhesion. Taken together, EMT in endometrial epithelial cells begins through stimulation of ovarian steroid hormones and is continued and amplified by the signals from the embryo.

After adhesion, it is required that embryos must invade and penetrate over the endometrial epithelial cell sheet for successful implantation; however, one must consider how embryos relocate from the luminal side of the endometrial epithelial cell sheet to the stromal side. Endometrial epithelial cells horizontally and tightly connect to neighboring cells and act as a physiological and immunological barrier; therefore, embryos cannot easily pass the barrier structured by the endometrial epithelial cell sheet. Since the late of 1990s, it has been observed that EEC apoptosis occurs at the embryo-EEC interface in mice [17], rats [18], monkeys [19], and humans [20, 21].

Destruction of the endometrial epithelial cell body by apoptosis is meaningful not only for the creation of the embryo penetration space but also because it provides additional effects. Apoptosis-induced release of self-derived components of destroyed cells includes nucleotides. Released nucleotides activate the immune system through Toll-like receptors. It has been demonstrated that the immune system is partly regulated by the interaction between UDP-glucose derived from the destroyed endometrial epithelial cells and its receptor P2RY14 expressed on the surface of the surrounding endometrial epithelial cells [22]. Given that interleukin-8 is simultaneously secreted from endometrial epithelial cells by UDP-glucose exposure [22], and that UDP-glucose appears at the site of implantation due to apoptosis, the immune cells recruited by interleukin-8 probably efficiently assist the implantation. In association with the apoptosis-induced creation of the embryo penetration route, loosening the connection among the endometrial epithelial sheet by EMT also benefits to embryonal intermittent penetration.

In fact, even if the apoptosis-initiating assistant system for embryo penetration is adequate, the characteristic dynamics of endometrial epithelial cells moving away from the site of implantation can be suitable for preparing an embryo penetration route more quickly than does apoptosis. At the initial step of implantation, endometrial epithelial cells may change the morphology of cells in preparation for apposition between embryo and maternal endometrium. Flattened and wide-spread morphological changes are observed in endometrial epithelial cells when treated with ovarian steroid hormones, such as estradiol and progesterone, or HDACI, suberoylanilide hydroxamic acid (SAHA) [15]. It is thought that the morphological changes in endometrial epithelial cells probably result from loss of cell polarity caused by EMT, and cells acquire motility for the following step of implantation.

Yamakoshi et al. have demonstrated that cadherin switch is observed in the trophectodermal cells [23]. Also in endometrial epithelial cells, cadherin switch has been demonstrated in in vitro implantation assay, and is enhanced by the stimulation of ovarian steroid hormones or SAHA. Interestingly, the significance of cadherin switch is accelerated by the adhesion of the embryo [16, 24] (Fig. 1). Similar to endometrial epithelial cells, endometrial stromal cells also migrate away from the position of implantation [25], and the orchestrated efferent migration of endometrial epithelial and stromal cells can aid the embryo penetration and invasion. Although no reports demonstrate that reversible MET occurs at the site of implantation after the embryo passes through the endometrial epithelial cell sheet, the sequential alteration of endometrial epithelial cells is probably recovered to a steady-state barrier structure by MET. In the regeneration of the endometrium, it is reported that MET is essential through menstruation, but not in implantation [9].

Clinical techniques in IVF-ET and EMT

At the clinical level, various trials are still being conducted due to the difficulties in improving implantation rates compared to the large improvements in fertilization techniques in in vitro fertilization and embryo transfer (IVF-ET).

To avoid multiple pregnancies, the stimulation of endometrium embryo transfer (SEET) method was developed to improve the pregnancy rate by injecting embryo culture fluid into the uterine cavity before embryo transfer instead of the two-stage embryo transfer method [26]. Embryo culture fluid containing stimulants which can probably induce EMT in endometrial epithelium, similar to the embryo.

Although a systematic review concluded that it did not improve production rates [27], multiple reports have shown that scratching the endometrium prior to embryo transfer contributes to improved pregnancy rates. It is thought that this effect is due to cytokines derived from destructed endometrium, probably containing UDP-glucose, as described above, and tissue repair after scratching likely requires EMT. At least, the set of scratching and remodeling of endometrium possibly assists in the initial phase of pregnancy, even if a significant improvement in live birth rates is not indicated.

Hormone replacement or additional therapy is the leading and necessary method of assisting in increasing implantation rates in IVF-ET. EMT-inducing factors also includes ovarian steroid hormones, as described above. HDACI can also induce differentiation and/or EMT in endometrial epithelial and stromal cells [12, 15, 16, 24]; however, one form of HDACI, valproic acid is prohibited for use in pregnant women because of its teratogenic properties. Although each HDACI has different genes as its controlling targets, they are also drugs that can control gene expression by as much as 2%, so their effect on embryos during the fragile period is a concern and care must be taken with their clinical application.

Systematic review and meta-analyses demonstrate that intrauterine injection of human chorionic gonadotropin (hCG) could be a valuable approach [28], and it is reported that hCG influence endometrial receptivity and EMT [29].

Conclusion

From the perspective of repeated regeneration, the importance of EMT in the endometrium, which is essential for tissue regeneration, can be understood by highlighting the large-scale regeneration of the human endometrium through menstruation and the local destruction and regeneration of implantation. In addition, dividing this into four stages helps us understand the relationship between the mechanism of implantation in the dynamics of endometrial epithelial cells and EMT. Revealing the mechanisms specific to tissue regeneration may provide the hint to solving problems related to the menstrual cycle and receptivity.

Acknowledgments

The author would like to thank Keio University for providing the experimental laboratory.

Disclosure

The author has any potential conflicts of interest associated with this research.

References
 
© The Japan Endocrine Society

This article is licensed under a Creative Commons [Attribution-NonCommercial-NoDerivatives 4.0 International] license.
https://creativecommons.org/licenses/by-nc-nd/4.0/deed.en
feedback
Top