The most valuable predictorsof endometrial receptivity

Summary

Objective:
To investigate the predictive factors of insufficient endometrial quality.

Design:
Review.

Setting:
Department of Obstetrics and Gynecology, General Faculty Hospital Prague, 1st Faculty of Medicine, Charles University.

Results:
This is the review on predictive factors of endometrial receptivity. There is a collected data on homeobox genes family, especially HOX 10 and HOX 11 members, that ensure implantation process by means of their interaction with estrogen and progesterone receptors via stimulatory proteins in the luminal epithelium in different periods of luteal phase. Moreover, the article describes appropriate secretion of osteoponin, cadherin, selectin, apopoliprotein D, mucin and balanced macrophage activity in endometrial epithelium in the proliferative phase of the menstrual cycle provide optimal conditions for implantation.

Conclusion:
Endometrial genes expressions along with the intrauterine environment macrophages activity maintain the tiny mechanism of endometrial receptivity.

Keywords:
implantation failure, endometrial receptivity, gene expression, osteoponin, integrin β3, apolipoprotein D, cadherins, selectins, HOX genes

INTRODUCTION

Over the last three decades despite multiple studies and meta-analysis there are unsolved questions on implantation failure, especially recurrent implantation failure in patients undergoing infertility treatment with the help of assisted reproductive techniques (ART). Endometrium quality as well as endometrium-embryo interaction form a group of keen interest among scientists on reproductive medicine all-over the world. This article presents the review on endometrium receptivity and other intrauterine pathology that may lead to implantation problems. This is the collected data on both animal models and human endometrium samples with a careful exploration of the receptivity and its influence upon the implantation rate.

In scope of general health there are some adverse factors of environmental pollution, life style and eating habits, obesity [14] that worsen the reproductive health. That is why the probability to conceive in one menstrual cycle for fertile couple even in optimal reproductive conditions is approximately 30% [27]. Recently, infertility has become the worldwide disease with the spread of about 9% among couples of reproductive age. It means, that almost one in 30 pregnancies in 2010 was achieved with the help of successful ART treatment (assisted reproductive techniques). Corresponding to ESHRE data over 4.3 million ART babies were delivered worldwide by 2010 [45]. It is hardly possible to understand the implantation failure without careful analysis of interaction between embryo and endometrial epithelium in uterine cavity.

ENDOMETRIAL RECEPTIVITY

In their works Cakmak et al. note that to achieve successful implantation embryo should prosper through several consecutive stages: apposition, adhesion and penetration of blastocyst to the luminal epithelium, that usually corresponding to the 5-6th day after the ovulation [2, 10, 11].

Moreover, if implantation occurs, the endometrium causes further molecular and morphological changes [2, 39]. At the final stage of implantation embryo invades through luminal into stromal epithelium for the purpose of interaction with maternal blood vessels necessary for appropriate development of trophoblast and embryo growth [2, 38]. Successful implantation is impossible without perfect endometrial receptivity.

HOMEOBOX GENES

Examination of the complex process of embryo – uterine collaboration is impossible without thorough understanding of what happens inside the uterine cavity, especially in both layers of endometrium, and what kind of substances are responsible for that. Upon that reason Taylor et al. determine an important role of homeobox (HOX) genes in embryo – endometrium interconnection. There are several reproductive homeobox genes on the X chromosome of humans, which are inherent in both sexes gametogenesis, implantation and embryo development.

First evidence about homeobox genes were performed almost 30 years ago by Lewis [26], in his works with Drosophila melanogaster. However, this group of genes has become a target of human reproduction since 2005, when new reproductive homeobox genes on X chromosome were detected. There are various numbers of HOX genes in different mammals, from 33 in mouse to 3 in humans. All HOX genes begin their numeration from centromere, from HOX1 to HOX13. Rodents HOX genes are divided into several clusters. Mice have α, β and γ clusters with four genes in each but some genes, like HOX 1, HOX6 and HOX 13 are not present in rats. This makes the principle species difference in HOX gene families.

The homeobox genes encode the transcription factors, containing homeodomain. Homeodomain is three α–spirals (helices) of 60 amino acids, which are bound with specific sequences on DNA as well as with specific proteins or RNAs [47].

Different HOX genes are expressed in different tissues and are responsible for different purposes.

The expression of HOX 10 and HOX 11 mRNA genes are increased in middle luteal phase of menstrual cycle and in case of successful implantation these genes expression continue to growth [2, 42]. Besides, they measured the expression of HOX 10 and HOX 11 mRNA genes in the endometrium of patients whose luteal phase was supported with medroxyprogesteronacetate (MPA) and 17β-estradiol [44]. It was determined that these genes were higher, comparatively to isolate usage of MPA or 17β-estradiol [2, 42]. Estrogens are usually considered to activate nuclear receptor (ER) that has two isoforms ERα and ER β. Estrogens regulate HOX 10 by means of ERα and ER β receptors. Both ERE1 and ERE2 (estrogen response elements 1 and 2 are detected in 5’ regulatory region of human HOX 10. In the estrogen receptors these elements interact with stimulatory protein 1 (SP1) binding site and CpG island. This mechanism activates the HOX 10 expression as a result of estrogen stimulation [42, 47].

On the other hand progesterone interacts via progesterone receptors (PR) that also exist in two isoforms PR-A and PR-B. In normal human endometrial epithelium, both PR-A and PR-B are increased by estrogens during the proliferative phase. Moreover, it was examined that PR-A predominates in stromal cells of endometrium and is responsible for progesterone-mediated stromal decidualization [2]. This may explain reasonable and adequate support of follicular and luteal phases of menstrual cycle, especially in the format of infertility treatment.

The corresponding data was obtained and proved at experimental model on mice. Foucher et al. detected that the implantation does not occur in the complete absence of HOX 10 and HOX 11mRNA genes. Besides HOX genes they have explored that the complex implantation process is regulated with β3 integrin and insulin-like growth factor-binding protein-1(IGFBP-1) [12]. In endometrial stromal cells of human beings HOX 10 is activated through the collaboration with FOXO (fork-head box proteins) transcription factor and together they up-regulate the expression of IGFBP-1(insulin-like growth factor binding proteins) [2, 12, 21].

These genetic factors interact via plasma membrane transformation. Garrido-Gόmez et al. have reported that pinopodes are generated by means of endocytosis of endometrial fluid from luminal epithelial cells. By that time the rest of endometrial cavity remains vacant. Furthermore, it may be optimal condition to generate the necessary mechanical contact of endometrium and blastocyst [25]. On the other hand further data proves the presence of pinopodes along the post- receptive period, so they could not be the reliable morphological marker of endometrial receptivity [13]. Along with other colleagues this group of scientists suggests to examine the epithelial secretion of mucin 1 and mucin 6 (MUC 1, MUC 6), integrins, leukemia inhibitory factor (LIF), calcitonin and HOXA 10 [7, 22, 24, 30, 37, 41]. This was also shown by Lessley et al., that biochemical markers are the best alternative to classic Noyes criteria to estimate the endometrial receptivity [25, 34]. It is important to mention that every phase of menstrual cycle demonstrates different gene expression.

Integrins are cell surface adhesion receptors, which increase during the implantation window. We may also observe changes in integrins expression at the beginning of pregnancy [29]. This expression is synchronized with trophoblast attachment αvβ1, α4β1, α6β1, α7β1 and embryo invasion into decidua is supported by α6β1, α5β1, α4β1, α1β1 integrins, which help to transform proliferative to endovascular extravillous cytotrophoblast [29]. The α 1, α v, and β 3 integrins are co-expressed in the epithelial endometrium only during days 19–23 of the menstrual cycle, supporting the concept that these cycle-dependent proteins, expressed within the putative implantation window, are potential biochemical markers of uterine receptivity. Moreover, some changes are present in osteopontin (OPN) expression – the principle ligand of integrin β3 which contains Arg-Gly-Asp binding sequences. OPN expression increases in glandular epithelial cells in secretive phase. However, there is a decreased OPN mRNA expression throughout the implantation window at endometriosis patients, that may explain the relatively worsen chances for implantation [46, 47]. Thus, enables us to classify the molecular status of endometrium according to its transcriptomic signature regardless to its histologic appearance [40].

There is a data on correlation of HOX genes expression and gynecological pathology. Patients with endometriosis present absence of HOX 10 and HOX 11 peak in secretory phase, and decreased HOX 10 expression due to hypermethylation of its promoter region [2]. There are similar problems in polycystic ovarian syndrome (PCOS) patients that present decrease expression of HOX 10, integrin β3, and IGFBP-1 in secretory phase [2].

INTRAUTERINE ENVIRONMENT

Salamonsen et al. have reported visual presentation of endometrium - embryo interaction that reflects knowledge about intrauterine environment that contains nutrients, enzymes, cytokines, antiproteases, transport proteins and biologically active factors. Luminal and glandular epithelium express emzymes, cytokins, antiproteases, transport proteins etc., and luminal epithelium has changes in mucin expression along with integrines in glycocalyx [15, 36].

Furthermore, in is important to mention that cadherins, immunoglobulins, and mucin are in a close interaction inside the luminal epithelium. Mucin, as a highly glycosylated glycoprotein is present on the surface of epithelial cells. Its gene locus is presented on the bond 21 of the long arm of the chromosome 1 (1q21). Mucin core proteins contain high levels of serine, threonine, alanine, glycine and proline residues. Estrogens stimulate MUC 1 and mRNA expression on the contrary to pure progesterone, which antagonizes only the stimulating effect of estrogens. This increases MUC1 and mRNA expression in luminal epithelium of humans are present during the secretory phase of menstrual cycle only [30].

Cadherins present a group of transmembrane glycoproteins that are responsible for calcium-dependant cell – to – cell adhesion mechanism. Intertalk of calcium and cadherin leads to adhesive mechanism, by means of triggering the cadherin redistribution that changes the polarity and adhesiveness of epithelial cells. Selectins play an important role in leukocyte transendothelial trafficking. Indeed, L-selectins are expressed on leukocytes and interact with their ligands on the endothelium that allows the rolling of leukocytes on inflamed vascular endothelium before their firm adhesion and transmigration. The L-selectin adhesion system is also demonstrated at the maternal–fetal interface, and may play a role in the initial steps of the implantation process [9].

Intercellular adhesion molecule-1 (ICAM-1) is a transmembrane glycoprotein that belongs to the immunoglobulin superfamily and is constitutively expressed on the cell surface. ICAM-1 mediates cell - to - cell adhesion since it constitutes a ligand for integrin molecules. In human endometrium, it is immunolocalized at the apical surface of the glandular and luminal endometrial epithelial cells as well as in the stroma throughout the menstrual cycle. Mucin-1 (MUC 1) and mucin-6 (MUC 6) were only found in thick glycocalyx that covers the apical surface of human endometrial epithelium [2]. Besides, MUC 1 does interfere with cell - to - cell and cell – to – matrix adhesion and is up-regulated by progesterone in a peri-implantaion period. However, scanning electron microscopy combined with immunohistochemistry demonstrated that MUC1 was only located on ciliated cells, and was missing from the surface of non-ciliated cells and pinopods. This suggests that pinopods may provide space free of the widespread MUC1 inhibition to embryo–endometrial interaction [2, 18]. LIF (leukemia inhibitor factor) has multiple biological activities in many different cell types, including proliferation, differentiation and cell survival - all functions that are essential for blastocyst deve-lopment and implantation [23].

UTERINE CAVITY GENES EXPRESSIONAND POSSIBLE REGULATIONOF EMBRYO – UTERINE INTERACTION

By means of high-density oligonucleotide microarray Cheong et al. have investigated the uterine cavity genes expression and possible regulation of embryo-uterine interaction. They have compared over 12,000 genes on the pre-receptive luteinizing hormone surge (LH +2) and receptive (LH +7) days of menstrual cycle in human endometrial biopsy samples (analyzed by an Affymetrix HG-95A array) [5]. In the variety of up and down regulated genes they have found only three genes being constant along the study. There were osteopontit (structural protein), apolipoprotein D (transporter) and Dickkopf / DKK1 – (signaling). The endometrial secretome is now known to represent a complex matrix that includes a number of mediators which modulate endometrial receptivity, and this may be involved in the maintenance and nurturing of ascending spermatozoa and preimplantation embryo. The primary components are proteins, amino acids, electrolytes, glucose, urea, cytokines, growth factors, metalloproteinases and their inhibitors, immunoglobulins, α-1 antitrypsin precursor, haptoglobin and transferrin [33].

However, other scientists have carefully examined the content of intrauterine fluid as the significant source of knowledge about glandular secretion that may play the crucial role in the implantation process. To prove their theoretical predictions they have inhibit secretion in glandular surface of the mice endometrium (mice have heamochorial type of placentation familiar to humans) [8, 20] and checked the implantation outcome. In humans, uterine fluid is multicomponent solution that consists of transudation of uterine vessels, glandular epithelium secretion, components of the tubal fluid. In conceptive cycle the uterine fluid compartment is moreover enriched by products of blastocyst´s secretion [36]. It is interesting to mentioned that despite the variability of gene expression along the menstrual cycle, patients with initial hormonal disorders (e.g. PCOS polycystic ovarian syndrome) or those who undergo ART treatment with controlled ovarian stimulation have emerged variable gene-based markers of the receptive phase [6, 17].

Cheong at al. have examined the predominant role of glycodelin (lipocaline superfamily glycoprotein with the presentation at long arm of chromosome 9), but not LIF (secreted glycoprotein with presentation at 22q12.2) in endometrial maturation. There was prior data on the influence of LIF secretion upon the placental grows. Besides, the soluble gel 130 form of LIF receptor is dramatically increased between LH +6 – LH +13 in endometrium of healthy women, in comparison to significant drop in infertile patients. Moreover, the LIF receptor mRNA was also detected in villous, extravillous trofoblast and endothelial cells of fetal villi, that proves their importance for adequate placentation [1]. In this samples, received with the help of endometrial fluid aspiration immediately prior the embryo transfer in IVF cycles (in vitro fertilization) 17 factors were examined. There were: interleukin 1β (IL-1β), IL-5, IL-6, IL-10, IL-12, IL-15, IL-17, IL-18, tumor necrosis factor α (TNF α), interferon β (IFN-β), macrophage migration inhibitory factor, exotaxin, IFN-β inducible 10 kd protein (IP-10), monocyte chemoattractant protein-1 (MCP-1), Dickkopf homolog 1, heparin-binding epidermal growth factor, and vascular endothelial growth factor (VEGF) [5]. It was detected that IL-1β and TNF α are mostly predictors of embryo quality rather than endometrial receptivity, but nevertheless they induce phosphorylation of transcription nuclear factor and translocate into nucleus. They bind to appropriate region of LIF gene inside nucleus that increases LIF in endometrium. Unfortunately, despite all scientific findings they have no predictive practical value due to complexity and expenses [15, 43].

Up to day there is still a question under debate about a value of (enzyme protein convertase) PC6 as a single marker of endometrial receptivity. It is secreted by human endometrial epithelium into the uterine cavity. Heng et al. have mentioned that PC6 decreased in the middle secretory phase of the menstrual cycle in some infertile women although, it remains insignificant to fertile women, so its specificity is under a judge [16].

Similar investigation on the endometrial quality along secretory phase was performed in 2000-s by Campbell et al. They have being performed by high resolution scanning microscopy with additional investigation of MUC-1, keratin sulphate and fucosilated lactosaminoglycans in epithelium gland fragment culture with antibodies (HMFG1, 5D4) and lectin (DBA – dolichos biflorus agglutinin). It was the contrary to latest data on endometrium gene expression and secretion of biologically active substances. They have noticed that keratin as well as DBA have dramatic increase through the secretarial phase of menstrual cycle, but it remains at highly heterogeneous distribution in glandular and luminal epithelium, and thus cannot be the significant predictor or even marker of endometrial epithelium quality [3].

Vargas et al. have performed investigation on the levonorgestrel influence on endometrial receptivity. The most important finding from review was an evidence of complete absence of histological changes in endometrial samples on LH+2 (luteinize hormone) and LH +7 days of menstrual cycle. At the same time, there were significant changed in gene expression (20383 genes were detected) at the same group, checked with the help of cDNA microanalysis and proved by real-time PRC (RT-PCR) [41]. These prove the mentioned above earlier findings on unreasonable usage of endometrial biopsy to detect endometrial receptivity [35].

MACROPHAGES INFLUENCE UPON GENE EXPRESSION IN EPITHELIAL ENDOMETRIUM

Moreover, interesting research on macrophages and their role in implantation process was noted by Nakamura et al. in 2012. They have explored the macrophage influence upon gene expression in epithelial endometrium and their links to glycan-mediated embryo adhesion in Ischiwaka, RL 95-2 and HEC 1. An endometrial epithelial cell culture was cultivated alone or with inactivated or lipopolysacharide-activated monocytic cells U937 [32]. U937 were separated using transwell inserts. However, getting into account the fact that a huge number of embryos are disable to implantate into the endometrium it should be mentioned that1/3(one-third) of embryos are failed to achieve the implantation due to incompetent yielding and 2/3 (two-thirds) are due to inadequate uterine receptivity or defects in communication between embryo and endometrial epithelium [4].

It was also examined that endometrium is richly populated with macrophages which are dependent on cytokins and chemokins. Cytokins and chemokins in their turn are regulated by ovarian steroids. Under these conditions macrophages have become potential regulators of molecules attachment in epithelial cells. Jasper et al. have presented data that in vitro fucosyltransferase-2 (Fut-2) is up regulated by macrophages or with the macrophage-derived mediators (LIF and interleukin 1β) [19]. This knowledge allows us to predict that tiny mechanism of maternal - embryo recognition during implantation may be regulated and thus improved by endometrial macrophages population. It is extremely important because macrophages are sensitive to infections, stress, inflammation, nutrition, smoking and environmental toxins. Their changes and condition can reasonable mediate adverse effect of these mentioned above factors on fertility [28]. The another mechanism of macrophageal influence upon fertility is the variety of their phenotypes and therefore interconnection with cytokine secretion profile, co-localization of CD 68 and secreted mediators (IL). It was also shown and proved by Miller et al. that endometriosis patients have reduced DBA (Dolichos biflorus agglutinin) bindings. This may be attributable to the altered numbers and potentially shift towards an inflamatory fenotype in population of their uterine macrophages [31].

CONCLUSIONS 

Endometrial genes expression along with intrauterine environment and macrophages activity maintain the tiny mechanism of endometrial receptivity. Although the definite predictive factors of endometrial quality are still being studies, it is obvious that both members of endometrium – embryo interaction team influence upon successful implantation.

Appropriate estrogen and gestagen supplementation in assisted reproduction treatment helps to prevent decrease in HOX genes expression and thus to save the endometrial receptivity. Thus embryo readiness to collaborate with receptive endometrial surface plays an important role at pregnancy onset. Low HOX 10 and HOX 11 expression along with suppressed secretion of macrophage activity cause implantation failure in infertile patients due to poor endometrial receptive. Despite integrins, mucins and LIF play plausible role in embryo – endometrial interaction none of them is an isolated predictor of endometrial receptivity in humans. The collected data on endometrial receptivity and blastocyst quality will be present in a different article.

MUDr. Tomáš Fait, Ph.D.

Gynekologicko-porodnická klinika1. LF UK a VFN

Apolinářská 18

128 53 Praha 2

e-mail: tfait@seznam.cz