Assessment of leukemia inhibitory factor and glycoprotein 130 expression in endometrium and uterine flushing: a possible diagnostic tool for impaired fertility

One in ten couples has problems conceiving. Of these, 25% have unexplained infertility [1]. Embryo implantation is a critical step in establishment of pregnancy. Implantation is the process by which the blastocyst becomes intimately connected with the maternal endometrium (decidua). In humans, the process of implantation can be divided into three phases: apposition, adhesion, and invasion. The opposition comprises the blastocyst orientation in the uterine cavity towards the endometrium. During the adhesive and invasive phases, the blastocyst approaches the epithelium, attaches itself to it, and the embryo trophoblast invades the deciduas. All these steps are controlled by a variety of interacting molecules of both maternal and embryonic origin [2]. Many cytokines, growth factors and adhesive molecules are known to trigger the initial process of implantation during the adhesive phase, and realize embryo–maternal contacts during the invasive phase [1].

The endometrium becomes receptive for a limited period of time after exposure to 17-β-estradiol (E) followed by progesterone. Embryo transfer studies have identified a phase of uterine receptivity, the “window of implantation,” between days 5 and 10 following the luteinizing hormone (LH) surge [3]. Many factors produced by the endometrium during this window have been proposed as molecular markers of a receptive endometrium. While the progesterone receptor (PR) is the only factor as yet identified to be absolutely required for successful implantation, leukemia inhibitory factor (LIF) and mucin 1 are regarded as two of the most important signaling vectors [4]. Leukemia inhibitory factor is a pleiotropic cytokine of the IL-6 family; i.e., it has effects on many different cell types and that its activities are not restricted to one lineage [5]. It is a polyfunctional, highly glycosylated protein (Mw 38–67 KDa) that mediates a wide range of effects in different ways, from promotion of cell proliferation and lifespan, to the control of their differentiation with regard to the tissue environment in which the target cells perform their functions. Mature LIF protein of maternal origin and its encoding mRNA have been identified in endometrium. Human fallopian tubes also express LIF; its mRNA is detectable throughout the cycle [6] and is expressed in cervical mucus [7]. In the endometrium, LIF expression remains low during the proliferative phase, rises after ovulation and remains high until the end of each menstrual cycle with its maximal expression during the mid-late secretory phase. Its high endometrial production in the middle and late phases of menstrual cycle stresses its important role in implantation [5].

Simultaneously, the blastocyst expresses LIF receptor (LIF-R), indicating the significance of LIF in embryo–maternal crosstalk. Endometrium and oocyst also express LIF-R [8]. In uterine flushing samples, LIF is measured to predict reproductive outcomes [9]. Other cells that express LIF-R include neurons, megakaryocytes, macrophages, adipocytes, hepatocytes, osteoblasts, myeloblasts, and kidney and breast epithelial cells. The LIF-R complex includes two subunits: the LIF-specific subunit LIFR-β, and the glycoprotein 130 (gp130) subunit (also used by IL-6 and IL-11) [8].

The gp130 subunit is a transmembrane protein. Soluble forms of gp130 are generated by proteolytic cleavage or by alternative splicing. Binding of LIF to its receptor promotes formation of a receptor complex with gp130. Signal transduction involves activation of members of the Janus family and phosphorylation of members of the STAT family of transcription factors. The LIFR-β/gp130 heterodimer can bind and signal in response to oncostatin M, ciliary neurotrophic factor, and cardiotrophin, in addition to LIF. Binding of LIF to LIF -R and gp130 activates signal transduction pathways [10].

The biological activity of LIF and IL-6 is affected by levels of their soluble (s) receptors: sIL-6R, sLIFR, and sgp130. The IL-6/sIL6R complex is able to bind and signal through membrane-bound gp130. Soluble gp130 acts as an antagonist—preventing the cytokine/receptor complex from initiating signals through membrane-bound gp130. In consequence, the ratio of LIF/sgp130 and IL-6/sgp130 is altered in endometrium of infertile women. Infertility in such cases appears to be associated with a failure by endometrium to express the normal molecular repertoire characteristic of the receptive period. Reduced sgp130 secretion may be a new molecular marker of such dysregulation in women with unexplained infertility [10]. In humans, LIF can be detected in samples from both endometrial biopsies and uterine flushing. The majority of women with unexplained infertility had dysregulated LIF production. Likewise, endometrial explants derived from infertile women showed reduced levels of LIF secretion [11]. Based on these finding, this study was designed to investigate the expression of LIF and its receptor subunit gp130 in endometrium of infertile women.

The implantation process is considered the most relevant limiting factor for successful pregnancy. Molecules that affect uterine receptivity for subsequent implantation of the blastocyst include integrins, colony stimulating factor-1 (CSF-1) and LIF. The fact that LIF is the key molecule in the implantation process in monkeys and mice inspired similar research in humans, which demonstrates that LIF mRNA is present in endometrium, and that increased LIF expression coincides with the implantation window. LIF belongs to the IL-6 family of cytokines, which signal through specific receptors on the cell surface that all share gp130 subunit as a common accessory signal transduction molecule [16].

The role of endometrial LIF in implantation has been documented in animals and human. LIF has been also found in the fallopian tube during preimplantation period, indicating a role for LIF in communication between the embryo and the tube. This observation suggests that absence of LIF in the mother, rather than the embryo, is responsible for failure of implantation [17, 18]. We started our study by confirming the expression of LIF mRNA in the endometrium of both fertile and infertile women during the implantation window. Expression of LIF mRNA in the endometrium was significantly reduced in infertile women compared with the fertile group. Dimitriadis et al. reported that immunostaining of endometrial tissue from fertile and infertile women for IL-11, IL-11Rα, and LIF —molecules affecting implantation—demonstrated equal staining for IL-11 and its receptor, but marked reduction of LIF staining in infertile women compared with the fertile group [19]. Mutations in the LIF gene have been described in nulligravid infertile women, and have been hypothesized to cause transcription abnormalities and decreased LIF expression [16, 20].

Measurement of LIF secretion in uterine flushing samples as a noninvasive technique enables the determination that low LIF concentration in such fluid during late luteal phase could be predictive of implantation failure [19]. The present study shows that secretion of LIF in samples of uterine flushing of infertile women during window of implantation was markedly low. Several studies have reported lower LIF in uterine flushing samples from women with primary unexplained infertility than in fertile women, and lower LIF secretion from endometrial explants of infertile women than fertile ones, especially during the implantation window [11, 19, 21]. However, these results contrast with those of Ledee-Bataille et al., who showed an inverse correlation between LIF concentrations in uterine flushing samples and the likelihood of successful implantation [22]. The conflict between the different reports might be due to the use of different ELISA kits. Reportedly, commercially available ELISA kits are far from equivalent, and mainly differ in their ability to detect glycosylated protein. Human LIF is known to be highly glycosylated; the sugar moiety represents as much as 50% of the total weight of the naturally produced protein. Several peptide epitopes may therefore be masked and the protein may not be detectable by the antibodies in certain kits [16].

While presence of sufficient amount of LIF protein seems to be an essential condition for implantation, a variety of other regulatory mechanisms also affect implantation, such as malfunction of LIF receptor, IL-6 signal transducer (gp130), which is an affinity modulator for the LIF protein receptor complex [16]. In a previous study, sgp130 was the most abundant member of the IL-6 cytokine-receptor family secreted by the endometrium at the time of implantation [10]. Therefore, the present study focuses on exploration of gp130 status in both endometrium and uterine flushing. The study showed minimal expression of gp130 splice variant1 in the endometrium of both fertile and infertile women equally. However, infertile women secreted significantly reduced levels of sgp130 in uterine flushing samples taken during the implantation period. Sherwin et al. demonstrated very weak gp130 immunostaining of endometrial tissue during the secretary phase of menstrual cycle [10]. Moreover, cultured endometrial biopsies taken from infertile women secreted significantly less gp130 than fertile biopsies did. Depending on that, the biological activity of LIF in the endometrium is affected by the levels of their soluble gp130 receptor. They reported that the reduced secretion of gp130 by infertile women during the widow of implantation points to a functional difference in endometrium at that time. Counter-intuitively, fertile women secrete higher soluble gp130 in uterine flushing samples in the presence of gp130 mRNA expression as low as that found in infertile women; this is apparently due to soluble gp130 being produced by proteolytic cleavage of the membrane-bound receptor.

In conclusion, the study indicates the possibility that endometrium of women with idiopathic infertility has abnormalities in expression and secretion of important cytokines as LIF and its gp130 receptor molecules, which may contribute to altered uterine receptivity and so infertility. The study also suggests that uterine flushing may help to detect the unreceptive uterus before assisted reproduction treatments and in the future, help to verify the normalization of cytokine concentrations to improve uterine receptivity before ovarian stimulation.