Abstract
MicroRNAs (miRNAs) have recently emerged as important regulators of gene expression stability. In the endometrium, miRNAs are involved in the dynamic changes associated with the menstrual cycle, implicated in implantation and in reproductive disorders. We performed a review in an attempt to assess the potential biological pathways linking altered miRNAs profiles with in vitro fertilisation (IVF) failure. Crucially, as miRNAs appear to have a significant role in the course of reproduction, they are excellent research candidates with the potential to enable a better understanding over the underlying molecular activities that prevent implantation and further progression of the embryo. Further steps include in-depth pathway mapping of the implantation process and the characterization of the respective miRNAs and associated links. The efficiency of any intervention should determine whether miRNA profiling could possibly be adopted in routine practice to substantially improve the diagnostic accuracy and, in parallel, the directed treatment of the next-generation IVF.
Although assisted human reproduction has evolved vastly over the past few decades marked by numerous milestone technological advances, some barriers are yet to be overcome. Among these is the control of the implantation process by fully understanding and successfully dealing with possible difficulties arising during this crucial step. Adequate endometrial receptivity and embryo development are essential for the pursued goal of increased pregnancy rates and reduced early pregnancy loss (1). Implantation failure is strongly related to the embryo transfer technique itself, embryo quality and endometrial receptivity and a synchronized dialogue between the maternal tissues and the blastocyst (2). In assisted reproductive techniques (ARTs), where the high-quality embryos are transferred, implantation remains the rate-limiting step for the success of treatment (3, 4). Therefore, a better understanding over the implantation process and the importance of the factors involved is warranted.
With the development of ‘omics’ technologies, numerous whole-genome expression analyses of the human endometrium have revealed hundreds of simultaneously up- and down-regulated genes that play a role in endometrial receptivity, embryo development and embryo-endometrial signaling (5, 6). Nevertheless, the molecular mechanisms regulating the expression of these genes are poorly understood.
In recent years, it has been demonstrated that small non-coding microRNAs (miRNAs) are important regulators of gene expression (7-9) and several miRNAs have been identified in the human endometrium (5). These small, single-stranded, non-protein-coding RNAs are a novel class of molecules with the ability to control gene expression at the post-transcriptional level through degradation, translational repression or silencing (10, 11). More than 2,000 different human miRNA species have so far been discovered in the human genome that may regulate the expression of one third of our genes and this number is steadily increasing (12). Every miRNA has a unique nucleotide sequence and expression pattern in a specific cell type (13-15). The biological transport of intercellular miRNAs and the current theories regarding the origin and the biological function of extracellular miRNAs are discussed in a recent review by Turcinovich et al. (15), while the fundamental approaches most commonly used for miRNA detection and target determination or regulation are discussed in recent reports (16-24). These studies point to their role, which is not yet fully elucidated, in key biological processes and physiological networks, including cell differentiation, proliferation, development and apoptosis. Consequently, and most importantly, their dysregulation is associated with the pathogenesis of various human disorders, such as infection, endometriosis, polycystic ovarian syndrome, recurrent miscarriage, subfertility and cancer.
Analyses of miRNAs in female human reproductive tissues have shown that both the uterus (endometrium, myometrium and cervix), as well as the ovaries, have high enrichment in individual miRNAs (22, 23, 25-27). The involvement of miRNAs in altered endometrial receptivity, abnormal pregnancies, endometriosis, gynaecological malignancies and fertility disorders has been reported (5, 28). Furthermore, in the ovary, events that reappear in a cyclic pattern, such as recruitment of growing follicles, atresia, ovulation and luteal tissue formation and regression, are believed to be regulated by genes controlled by miRNAs (26); similarly, miRNAs play a semantic role in the regulation of the oocyte and cumulus cells crosstalk (27).
Since their role in modifying gene expression is now well-established, an overview of the level of their involvement in human reproduction could possibly provide new insights into the management of implantation failure in in vitro fertilisation (IVF). The purpose of the present review was to assess the potential biological pathways linking altered miRNA profiles and IVF failure using the most up-to-the-minute available data through detailed investigation of the studies addressing this issue and the respective references.
Search Strategy and Results
This review was conducted according to an initial protocol agreed by all authors. Studies in English language from 1976 to July 2014 were compiled with no study design restrictions using the following search algorithms in two major scientific/medical databases (Medline/PubMed, ScienceDirect): (“miRNA” OR “microRNA”) AND (“IVF” OR “in vitro fertilization” OR “in vitro fertilisation” OR “assisted reproduction” OR “Reproductive Techniques, Assisted” OR “ART” OR “reproduction” OR “implantation failure” OR “IVF failure”). Search results were confirmed through the SCIRUS database. Reference lists of relevant articles were hand-searched for potentially eligible studies.
Data were retrieved from 80 out of the 248 eligible titles found at the initial search. The main text of the review was divided in 3 parts; data from experimental models, the alterations of miRNAs during controlled ovarian hyperstimulation (COH) for IVF and the association of these alterations with the procedure of human implantation. The potential pathways retrieved from the sum of the studies are presented in Figure 1.
Animal Studies
miRNA-mediated regulation of uterine gene expression in the context of implantation has been demonstrated in animal models with findings that confirm and underline the necessity of the dialogue between maternal and embryonic tissues (29, 30). Nothnick et al. (22), in a comprehensive review, included studies demonstrating that miRNAs are essential for normal development and function of the female mice reproductive tract, for example through the conditional deletion of the miRNA processing enzyme Dicer1 and an miRNA expression profile linked with mice reproductive pathology. Paradigms on the involvement of differentiated miRNAs in implantation in animals are numerous.
Cui et al. (31) and Murchison et al. (32) showed that Dicer1 was highly expressed in oocytes, with the highest-level expression being in the transcriptionally repressed germinal vesicle oocyte. Dicer1−/− mice were infertile, although their ovaries were histologically-normal and responsive to gonadotropins due to the association of the changes in the spindles, which were misaligned and disorganized, resulting in a loss of meiotic maturation and polar body production. Microarray analysis of these Dicer1−/− oocytes showed that 861 mRNAs were up-regulated >2-fold, with 173 of these having previously been shown to be degraded during meiotic maturation (32). Just recently, Zhang et al. reported the first microRNA-target mRNA pair (mmu-miR-376a modulating the expression of Pcna) that regulates the primordial follicle assembly in mice (33).
Moreover, dre-miR-430 was found to bind to over 300 maternal transcripts in zebrafish promoting their down-regulation directly before the beginning of the embryonic transcription (34), while mmu-miR-101a and mmu-miR-199a* were reported to be spatiotemporally expressed in the mouse uterus during implantation and post-transcriptionally regulate the expression of cyclooxygenase-2 (29). bta-miR-122a and * bta-miR-199a* were specifically expressed in day 30 bovine embryos (35) targeting cyclin G1 and extracellular signal-regulated kinase 2, respectively (36, 37). Accordingly, ssc-miR-181a and ssc-miR-181c play significant roles in the regulation of genes and pathways involved in embryo implantation and placentation in pigs (38).
Also, up-regulation of rno-miR-98 in ratus uterus during the receptive phase is linked to the decrease of endometrial stromal cell (ESC) proliferation via targeting B-cell lymphoma–extra large gene (Bcl-xl) (39).
Wang et al. (40) extracted total RNA from 2-cell and 4-cell mice embryos and observed that 192 miRNAs were differentially expressed between these two (122 up-regulated and 70 down-regulated), thus providing a developmental map for a large number of miRNAs; the authors' results obtained through microarray analysis were further confirmed by real-time quantitative RT-PCR for six miRNAs (mmu-miR-467h, mmu-miR-466d-3p, mmu-miR-292-5p, mmu-miR-154, mmu-miR-2145 and mmu-miR-706), all appearing to exert a special, but not yet elucidated role, in reproduction. Interestingly, a differentiation of miRNAs has been proven between implantation and inter-implantation sites (41).
miRNA levels were found to be different in IVF porcine embryos and in vivo and were related to the stage of embryonic development and to culture conditions (42). Tang et al. (43) found that between the 1-cell and 2-cell stage there was a 60% decrease in miRNA expression levels indicating that miRNA may be degraded similarly to maternal mRNA. However, between the 2-cell and 4-cell stage, as embryonic transcription proceeds, miRNA expression increases ~2.2-fold. Those from the miR-290 cluster have previously been shown to play a role in DNA methylation and telomere-length homeostasis (43, 44).
Alterations of miRNAs During COH
Several studies have been published on the physiological endometrium in an attempt to determine its gene profile (45-48). In the endometria of sub-fertile women, genetic analysis has demonstrated dysregulation of the respective pathways linked with leukocyte extravasation signalling, lipid metabolism and detoxification (49). Gene expression of endometrial cells is mainly under the influence of ovarian steroids and in COH, during IVF, ovarian steroid secretion is elevated (50), with estradiol levels increasing by 5-fold (51) or higher.
The basic concept of miRNAs' involvement in implantation lies on the modulation of endometrial and function / receptivity (52). In IVF, COH itself alters the gene expression profile through alteration of the mRNAs profile, when compared with natural cycles (53). In contrast, during the menstrual cycle of fertile women, there is no association between the changes of the female body with the expression of the circulating miRNAs (54).
Receptivity is negatively affected by the COH itself (28, 55, 56). This is accomplished through structural and functional alterations, which include promotion of histological functions and of pinopode maturation, as well as down-regulation of steroid receptors (53). During COH, along with a “genomic delay’ of the endometrium, there is a localized expression of a unique set of miRNAs, in both the endometrium and the ovary, which is found to be altered, especially during the implantation period in the former (57, 58). This set of miRNAs is diverse when compared to that of natural cycles as shown by the deviant numbers and species found to be down-regulated or, more rarely, up-regulated. As a result, the necessary function, and especially regression of specific target gene expression, is altered at a cellular level. It is well-accepted that if critical genes are not expressed during the window of implantation (WOI), implantation will not happen (59). It seems that during COH, an atypical pathway is followed further involving genes responsible for anti-inflammatory and anti-apoptotic roles and those implicated in angiogenesis (60, 61).
In addition, most reports conclude that the alterations during COH can consequently alter the WOI due to the abnormal hormonal environment to which the endometrium is exposed (50, 62-64). Interestingly, a recent report has provided a link between the type of ovulation trigger and luteal phase support during COH for IVF and endometrial receptivity and early implantation of the embryo (57). Of note, at the site of implantation, endometrial gene expression differs between fertile women and women with repeated implantation failure during IVF or recurrent miscarriages. Lédée et al. (65) performed a large-scale microarray analysis to provide evidence of an extensive endometrial dysregulation in these patients, which seems to reflect the impaired ability of the endometrium to co-act with the embryo at the time of implantation. Moreover, high levels of steroids themselves promote variations in the concentration of cytokines, adhesion molecules, growth factors and enzymes, such as leukemia inhibitory- and heparin-binding epidermal growth factor (EGF)-like growth factors, interleukins 4, 6, 10, 11, 13, 15 and tumor necrosis factor-α (TNF-α) (66) and in the physiological angiogenesis in the placental site (67).
These events appear to lead to the observed asynchrony between the endometrium and the embryo stage, observed in repeated IVF implantation failures following COH (68, 69). According to recent data, the alteration of specific subsets of miRNAs results in the differentiated expression of, for example, cytokines, such as osteopontin (OPN) and vascular endothelial growth factor (VEGF) and the inability to stabilize the endometrial expression of specific genes, which, in turn, lead to the decreased endometrial receptivity and subsequently poor pregnancy rates in IVF patients (60, 70).
In conclusion, although there is no direct evidence indicating that altered miRNAs' expression results in the differentiated expression of these genes and abnormal expression of miRNAs is just one of potential reasons for altering genes expression, it seems that continuous research will eventually firmly link the pieces of the puzzle.
Embryo Implantation
During the normal procedure of implantation and maintenance of pregnancy, a precise series of molecular interactions between the blastocyst and endometrium is necessary. Both biological processes are based essentially on adequate gene expression that, in turn, depends on miRNAs function. The conditions under which this molecular network functions determine the fate of the blastocyst itself. Unfortunately, it is not easy to provide an exemplified pathway to assist our understanding over the underlying functional networks.
In subfertility, there is a marked difference in the miRNA profiles when compared to fertile population. This difference seems to be age-related as described through the detection of miRNAs in membrane-enclosed vesicles of human follicular fluid (71). There also has been observed a significant decrease in the expression of six miRNAs in comparison with fertile donor control blastocysts derived from patients with polycystic ovaries or male factor sub-fertility while being morphologically similar (72). Analysis of the target genes demonstrated significant altered expression consistent with aberrant miRNA profiles (72). As a result, there is an etiological derivation in major biological processes, such as in cell growth and maintenance, transcription and molecular functions implicated in nucleic acid binding and signal transducer activity.
In the same context, differential miRNA expression between euploid and aneuploid embryos has also been reported to be a potentially early indicator of their prognosis or a mechanism behind their eventual fate (73). Similarly, Rosenbluth et al. found that miRNAs (miR-645, -191 and - 372) into culture media of embryos at day 5 positively correlated with embryonic aneuploidy and pregnancy outcomes (74).
Assou et al. (27) using a deep-sequencing approach studied the identification and quantification of miRNAs in human MII oocytes and cumulus cells (CCs). The authors report that BCL2 and CYP19A1 mRNA levels were decreased upon MIR23a over-expression concluding that miRNA could play a role in the regulation of the oocyte and CC crosstalk (27). Li et al. (70) performed a gene pathway analysis of miRNA and mRNAs examining the effects of a differential expression between normal and elevated progesterone (P) in women undergoing IVF: the analysis identified OPN and angiogenin and reduced VEGF, as playing an important part in impaired endometrial receptivity in cases of elevated P, leading to poor pregnancy rates. The authors demonstrated that several of the differentially-expressed genes have been identified as target genes for miRNA, which are believed to play a crucial role in reducing endometrial receptivity. Furthermore, they identified four miRNAs and 22 mRNAs that were differentially expressed between normal and elevated P endometrium. Of note, the studies assessing endometrial gene expression have included only a small number of patients due to the cost, on one hand, and the plethora of the involving genes, on the other. In addition, the results reported are quite different through possibly different assays and methods used; accordingly, the conclusions that can be drawn are limited (28).
In the special group of cases with repeated implantation failures, a reduced action of both miRNAs and the targeted molecular pathways has been demonstrated as compared to those of fertile women (21, 75). Possible causes were attributed to inappropriate timing, the lack of expression or the asynchronization between miRNAs and the gene target expression itself. As a consequence, the differentiated miRNA-regulated molecular pathways lead to defects in endometrial receptivity and, hence, to unsuccessful embryo implantation.
Conclusions and Next Targets
MicroRNAs are involved in numerous processes affecting the function of the human reproductive system and, moreover, conception, implantation process and embryonic development. Crucially, as they play a significant role in the course of reproduction, they represent excellent research candidates with the potential to enable a better understanding over the underlying molecular activities that prevent implantation and embryo progression. During an IVF cycle, COH triggers differential miRNA expression and, since these are responsible for the suppression of target gene expression, this has an adverse effect on the implantation process. Recent research allowed for the characterization of specific genes whose expression is associated with implantation and their functional interplay with miRNAs. A schematic summary of the potential pathways linking altered miRNAs profiles and IVF failure is presented in Figure 1.
A vital step is to support the continuity of current research for in-depth pathway mapping of the implantation process and the characterization of the respective miRNAs and associated links so as to unveil possible new roles of both previously and newly discovered molecules. More specifically, there is an urgent need for the identification, validation and functional characterization of specific miRNAs involved in implantation along with parallel assessment of their targets in order to clarify their role in fertility and implantation. To this end, various novel miRNA-associated molecules in the form of agonists / mimics or antagonists and synthetic decoy targets have been put forward. Advanced techniques and methodologies have already been proposed for combining miRNAs by integrating different co-expression networks in order to be accurately utilized in biological models. Since a single miRNA can potentially affect multiple targets, an indisputable point here is that, before a proposed protocol is applied in humans, any potential intervention in these pathways for therapeutic purposes should first be monitored in variant models to weigh its positive against any adverse effects.
In addition, relevant data from gene expression studies should be integrated to find markers of clinical value. Thus, targeting specific miRNAs can potentially represent a valuable approach in the directed treatment of specific cases where multiple factors affect fertility and result in IVF failure. Moreover, precise characterization could possibly uncover an important marker of fertility status in both sexes or even markers for monitoring, analyzing and potentially improving gamete quality and endometrial receptivity.
Cost effectiveness should be explored and, furthermore, trials on the efficiency of any intervention should determine whether miRNA profiling could possibly be adopted in routine practice to substantially improve the diagnostic accuracy and, in parallel, the directed treatment of the next generation IVF. Cost reduction to a minimum of any potential future application should be seriously considered so that such services may be accessible to all groups of patients. Up-to-date published data and ongoing trials are limited, probably attributable to the fact that cost could place a series of obstacles in performing large-scale studies for the purpose of establishing robust conclusions on the effectiveness of this practice.
Acknowledgements
The Authors would like to thank Dr. Aggeliki Kolialexi, Department of Medical Genetics, Athens University School of Medicine, for her valuable input in scientifically editing the text.
- Received November 23, 2014.
- Revision received December 2, 2014.
- Accepted December 8, 2014.
- Copyright © 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved