**3. Endometrial receptivity in women with polycystic ovary syndrome**

Endometrial receptivity is a complex feature of the endometrium that allow the embryo to attach and invade the endometrium, and its further development into a viable fetus. For normal implantation both embryo and endometrial quality are important. In normal women the endometrium is receptive to embryo implantation for a period of 3–6 days which starts seven to ten days after ovulation known as the window of implantation. In pathological conditions this window can be shorten or shifted, resulting in infertility or pregnancy loss.

Human endometrium is a tissue whose development depends on the level of circulating hormones. During the follicular phase of the cycle, the increasing circulating estradiol levels determine the proliferation of the endometrial cells and increased endometrial sensitivity to estrogen through increasing the estrogen receptors (ER) ER-α levels [24, 25]. The ER expression is highest in the late proliferative phase, decreasing in the luteal phase [25]. Following ovulation, the progesterone production determines inhibition of cellular proliferation, mitotic activity, DNA synthesis and stimulates the differentiation of the endometrial cells [26]. The inhibition of proliferation of the epithelial endometrial cells is the consequence of the progesterone-induced regulation of genes resulting in down regulation of estrogen receptors and the induction of the enzymes that metabolize estrogens reducing its cellular effects [26]. Moreover, progesterone reduces the expression of androgen receptor in endometrial cells and stroma [26]. As a consequence of all these changes the 'window of implantation' occur. Progesterone is also essential for decidualization, a process that allow trophoblast invasion in case the implantation occur and establish a cytokine milieu and immunomodulatory network in the stroma. The decidualization is the consequence of endometrial stromal cells modifications of the cytoskeleton and up-regulation of prolactin, insulin-like growth factors, IGF binding proteins, insulin receptor and other factors. In the case the implantation does not occur, the decreasing estrogen and progesterone levels determines a shift from the expression of the innate immune genes to inflammatory genes expression in association with cellular apoptosis, increased production of metalloproteinases and prostaglandins, followed by endometrial desquamation and menstruation [26].

In women with PCOS, in the absence of ovulation the progesterone effects on the endometrium are lacking or severely decreased, affecting the decidualization and the window of implantation. Moreover, women with PCOS may have increased exposure to estrogen levels [27] as a consequence of aromatization of increased androgens in adipose tissue and decreased sex hormone binding globulin due to hyperinsulinemia [28, 29]. Several experimental studies support the alteration of endometrium especially in the window of implantation. Thus, Avellaira et al. [30] found that the tissue homeostasis in secretory endometrium of untreated women with PCOS is affected by an imbalance between apoptosis and cell proliferation which is increased as demonstrated by a study evaluating the expression of the proteins related with the two processes [30].

It was also suggested that some endometrial alterations in PCOS are the consequence of prenatal intrauterine exposure to androgens being considered primary endometrial abnormalities [23]. Thus, the endometrium of PCOS women has a preponderance of estrogen and androgen action and decreased progesterone action as a consequence of hormone receptors expression and function.

#### **3.1 Estrogen receptors (ER) expression and function**

In PCOS patients the level of ER seems to be increased in all the cycle phases [31–35]. Moreover, some coactivators of the ER-α like TIF2 and AIB1 were also

found in higher levels in the proliferative endometrium of the PCOS patients compared with controls [32–35]. Moreover, during the proliferative phase, the endometrium of PCOS women showed a higher Bcl-2/Bax ratio, indicating the predominance of anti-apoptotic factors in the estrogen receptor increased environment [31]. The higher levels of mRNA and protein for ER-α and coactivators compared with normal women were reported in the mid-secretory phase endometrium of PCOS as well [32]. Moreover, the coactivator ARA70 was increase and epithelial expression of beta3-integrin, a protein involved in cell adhesion and cell surphace mediated signaling, was decreased in endometrium of PCOS versus control [32].

Enhanced ER-α activation was associated with rare or absent apoptosis and increased in cell proliferation in the endometrium in the mid luteal phase [30, 36–38]. On the other hand, it is possible that high circulating estrogens as in patients performing ovarian stimulation to determine the apoptosis of the endometrial glandular cells, as showed by an experimental study by Chen et al. [39].

Other factors can contribute to increased exposure of endometrial cells to estrogens. Thus, the modified activity of the endometrial enzymes 17-β-hydroxysteroid dehydrogenase, which is downregulated and hydroxysteroid dehydrogenase type 1, which is upregulated, could contribute to increased local production of estradiol and androstenediol with increased estrogenic activity [40–42].

#### **3.2 Progesterone receptor (PR) expression and function**

In endometrium of PCOS women progesterone receptors (PR) alpha and beta mRNA is overexpressed and the stromal immunostaining of PR-b and Ki67 is higher compared with BMI-matched controls [43]. Moreover, an imbalance between PR-a expression and PR-b was found in proliferative endometrium of obese PCOS women, with the predominance of the later [43]. It was also reported that endometrial tissue of PCOS patients has decreased responsiveness to progesterone [44], also known as 'P resistance' [45]. This progesterone resistance might be due to reduced binding and activation of PR [46] or to an altered expression of its isoforms [46, 47]. Elevated PR isoform expression was showed to be associated with increased systemic levels of estrogen [48] and of androgen [49, 50]. Thus, hyperandrogenism modulates the expression and function of PR being associated with inactive or less active isoforms of PR [49–51], being probably responsible for progesterone resistance in PCOS women.

Mucin 1 (MUC1) is a progesterone-regulated molecule that carries selectin ligands recognized by the human blastocyst. Thus, Margarit et al. [52] showed that MUC1 expression is lower in anovulatory PCOS than in fertile patients, being a possible contributor to decreased implantation.

#### **3.3 Androgen receptor**

In the endometrium of PCOS women the androgen receptor (AR) mRNA and protein expression is increased [23] and coactivators of AR were found to be overexpressed as well [37]. Among these coactivators, Melanoma-associated antigen 11 (MAGEA11) was showed to bind to AR resulting in alteration of window of implantation [11]. Endometrial microenvironment can also contribute to over-exposure to androgen. Thus, low level of SHBG can increase the bioavailable testosteron at endometrial level [53] and increased activity of endometrial 5a-reductase generates potent androgens such as di-hydro-testosterone [37]. The uptake of intracrine precursors of testosterone [38, 54] and increased activity of hydroxi-steroiddehydrogenase might contribute to high androgen exposure of endometrium in PCOS [55].

## **3.4 Hyperandrogenism**

During the mid-secretory phase of the menstrual cycle and the window of implantation the level of circulating androgens reach a nadir in normal cycles [56], while in PCOS patients with hyperandrogenism the level of circulating androgens is constantly increased. There is a significant body of evidence suggesting that overexposure to androgens may affect the normal development of endometrium and, therefore the endometrial receptivity. Thus, Cermik et al. [57] studied the effect of testosterone on HOXA10 expression in endometrium, a gene well known to be involved in endometrial receptivity. They demonstrated that, in vitro testosterone exposure decreases the expression on HOXA10 and prevents the increase of this gene in response to estrogen and progesterone [57]. They also confirmed that the expression of HOXA10 is decreased in the endometrium of the hyperandrogenic PCOS patients [57]. Homebox (HOX) are genes essential for endometrial receptivity which are maximally expressed in endometrium during the window of implantation [58]. Their importance in implantation is demonstrated by experimental studies on female mice with disruption of HOXA10 which show infertility with implantation failure in spite of the presence of ovulation [59].

It was also showed that Wilms tumor suppressor (WT1) gene which is expressed in endometrium in the window of implantation, was downregulated in ovulatory women with PCOS in comparison with normal controls and that this downregulation is the consequence of androgen exposure in in vitro models [60] Since high androgen are associated with elevated PR isoforms, it was suggested that hyperandogenism could be a contributor to progesterone resistance found in PCOS women [50, 61]. High androgens can also affect the number and function of endometrial pinopodes which are associated with endometrial receptivity [62, 63]. Androgens can also influence the decidualization by their ability to modulate the oxidative stress response in decidualized endometrial cells [64] since the oxidative stress was showed to influence factors involved in embryo implantation like cytokeratin 8 (CK-8) [65]. Other androgens can also act on endometrium. Thus, DHEA seems to block glucose utilization resulting in inhibition of decidualization [66] and modulation of cell survival and apoptosis [67]. In pregnant mice treated with DHEA an impaired LIF-signal transducer and activator of transcription 3 (STAT3) pathway was observed, which was associated with implantation failure [68].

Clinical data showed that pregnant women with PCOS have lower endovascular trophoblast invasion in relation to circulating testosterone and the clinical phenotypes involving hyperandrogenism [69, 70].

#### **3.5 Hyperinsulinemia and insulin resistance**

In clinical studies, in PCOS patients undergoing in vitro maturation-in vitro fertilization embryo transfer cycle insulin resistance was associated with decreased implantation, clinical pregnancy and ongoing pregnancy rates [71]. Energy metabolism is vital for proper endometrial function taking into consideration the rapid turnover of endometrial tissue. Therefore, insulin action on endometrial tissue might be essential for endometrial receptivity. Insulin resistance and compensatory hyperinsulinemia can have detrimental effects on the implantation process as hyperinsulinemia has been shown to impair stromal cell decidualization in vitro [72–74]. In PCOS endometrium several abnormalities in insulin signaling and glucose transport have been reported. Thus, hyperinsulinemia can reduce insulin receptor substrate 1 (InRS-1) activation and glucose transport in endometrial stromal cells in PCOS women [75–76], probably via inflammatory pathways [76, 77]. Experimental studies showed that in vitro exposure to dihydrotestosterone altered the expression

of insulin receptor and insulin receptor substrates and the phosphorylation of insulin receptor in endometrial stromal cells [74]. Other studies suggests that insulin action is decreased in endometrium of hyperinsulinemic women with PCOS, by showing that pAS160T642 and SLC2A4 which are substrates of insulin receptor are decreased in comparison with non-hyperinsulinemic PCOS and controls [75].

In vitro studies showed that insulin inhibits the production of IGFBP-1 in the endometrial stoma. IGFBP-1 is considered a biomarker of decidualization [73]. It is also possible that adiponectin, an insulin regulating molecule and regulator of glucose metabolism, which is decreased in PCOS patients, to be involved in endometrial receptivity of these patients. Thus, adiponectin receptors were found to be highly expressed in the human endometrium during the window of implantation [78]. Moreover, decidualized mouse endometrium is able to secrete adiponectin and adiponectin receptors were found both in decidual cells and embryo after implantation [79]. This data supports the hypothesis that adiponectin might play a role in endometrial receptivity and implantation. Studies showing that metformin reduces the miscarriage rates in PCOS patients indirectly support the hypothesis of insulin resistance involvement in endometrial receptivity [80].

#### **3.6 Vitamin D**

Vitamin D deficiency is a frequent condition among women with infertility or PCOS. Numerous studies suggested a role of vitamin D in reproductive health at variate level of reproductive system, including endometrium. However, clinical data regarding the association between serum vitamin D level and endometrial pathology are divergent. A prospective controlled study [81] analyzed factors associated with recurrent implantation failure and found higher prevalence of chronic endometritis, a lower vitamin D level and a borderline lower progesterone level in comparison with controls [81]. However, two recent systematic reviews found no association between serum vitamin D level and miscarriage rate in women who performed in vitro fertilization [82, 83], while one of these reviews reported higher pregnancy rates in vitamin D replete patients undergoing assisted reproduction treatments [82].

Experimental studies support the role of vitamin D in endometrial receptivity. Thus, Guo et al. [84] found that circulating level of vitamin D was positively associated with vitamin D receptor and HOXA10 protein level expression in the endometrium and these were substantially elevated in pregnant women compared to non-pregnant women. During the window of implantation, higher serum vitamin D levels were associated with more mature pinopodes (84). Another study showed that in patients with recurrent implantation failure vitamin D treatment increases progesterone receptor mRNA and protein level and its phosphorylation on Ser294 residues in the endometrial cells (85). These results suggest that vitamin D may play a key role in the endometrial receptivity [85].

In patients with PCOS vitamin D deficiency is frequently found [86]. However, specific data regarding the relationship between serum vitamin D and endometrium in PCOS patients are scarce. Several studies reported that vitamin D treatment can improve the features associated with endometrial receptivity like hyperandrogenism [87], hyperinsulinism and insulin resistance [88], inflammation and oxidative stress [89].

In an experimental study in a rat model of PCOS it was found that immunohistochemical staining of caspase-3 and Ki-67 were decreased with vitamin D treatment compared non-treated group [90]. Moreover, endometrial, epithelial and stromal thickness measurements were decreased in the vitamin D treatment group compared to non-treated PCOS group [90].
