The Possibility of Analyzing Endometrial Receptivity Using Cells from Embryo Transfer Catheters

It is very important to investigate the expression of endometrial receptive markers in the endometrium during implantation. Therefore, we examined whether it would be possible to analyze endometrial receptivity using cells from embryo transfer catheters. A total of 81 cycles from 81 consenting patients were enrolled in this study. The tip of the embryo transfer (ET) catheter was cut and immersed in a dedicated reagent. Confirmation of cell distribution was carried out using a Papanicolaou stain and immunocytochemistry. Protein expression was carried out by immunocytochemistry. The expressions of estrogen receptor α, progesterone receptor, and homeobox A10 mRNA were analyzed using quantitative reverse transcription-polymerase chain reaction. We analyzed the relationship between the gene expression profiles associated with pregnancy from endometrial cells. Samples collected from the ET catheter showed clear staining for endometrial cells. Most of the cells were endometrial epithelial cells. Cervical cells were not observed. The protein expression was also confirmed. Three genes were analyzed that are associated with endometrial receptivity. Progesterone receptor expression was 1.4-fold (p<0.05) and homeobox A10 was 2.8-fold (p<0.01) higher in patients who became non-pregnant group, compared to the pregnant group. Estrogen receptor α expression tended to be higher in the non-pregnant group (p=0.18). Our results suggest that endometrial receptivity can be evaluated using cells obtained from the ET catheter. This method may be useful for elucidating the cause of implantation failure by comparing a receptive and non-receptive endometrium at the time of ET.


Introduction
Implantation of the blastocyst into the uterine endometrium remains a significant limiting step for success in in vitro fertilization and embryo transfer (IVF-ET) [1]. It is important to gain an understanding of the fertile endometrial physiology that can successfully support implantation, and the subsequent maintenance of pregnancy and placentation. We previously reported that endometrial cell function is changed by decidualization in association with increasing proteaseactivated receptor (PAR)-1 expression [2]. The upregulation of PAR-1 in decidualized endometrial stromal cells may have some effect on pregnancy.
In human endometrial epithelial cells, gene expression of cytokeratin (CK) 15 is reported to decrease in a homeobox A (HOXA) 10-dependent fashion [3]. It is reported that dramatic changes in cellular architecture are necessary to achieve the secretory changes in the endometrial epithelium that bring about the implantation window. Even euploid, morphologically normal blastocysts fail to implant in about one-third of transfers [4,5]. This failure rate for the implantation of euploid embryos may suggest a non-embryonic source of implantation failure, with endometrial receptivity issues representing another potential cause. It has been recognized that, for successful implantation, there must be developmental synchrony between the embryo and the endometrium [6].
Recently, the endometrial receptivity array (ERA) was developed as an objective molecular dating method to accurately and reproducibly identify endometrial receptivity status from endometrial tissue samples [7]. There have been several reports concerning the effectiveness and significance of the ERA [8][9][10]. One limitation of endometrial gene expression studies has been the need to use cells collected by endometrial biopsy for profiling in the previous cycle [11]. Sequential invasive sampling of the endometrium during a single cycle can result in wound-related confounding and can alter the biomarker candidates. Temporal and regional gene expression variations found within the endometrium may make selecting the ideal conditions for embryo implantation difficult. It has been reported that several factors affecting implantation expressed in proliferative endometrial cells obtained by biopsy during the ET cycle [12]. However, we think that a more direct analysis of the status of the endometrium will be needed. Therefore, we tried to collect endometrial cells in a minimally invasive way from the implantation site of an ET cycle. The purpose of this study is to clarify the possibility to analyze endometrial receptivity using these cells from an ET catheter.

Patients
A total of 81cycles were studied from 81 consenting patients enrolled in this study who were treated for infertility at St. Luke Clinic during the period from December 2019 to June 2020. Sequential hormone replacement therapy (HRT) was employed to prepare endometrial receptivity in freezethawed blastocyst transfer cycles. Informed consent was obtained from all patients in accordance with a protocol reviewer and approved by the institutional ethical review board of Oita University, Yufu, Japan.

Endometrial Cell Isolation
ET was performed on day 5 with confirmation of uterine conformation as well as evaluation of the endometrium over 7 mm using transvaginal ultrasonography. At that point, progesterone had already been used for 5 days. Once these conditions were identified, a flexible guide for the ET catheter (KITAZATO, Fuji, Japan) was introduced through the cervix, ensuring contact with the endometrium with ultrasonography guidance. After the guide position was determined, the stylet in the guide was removed. The ET catheter through the guide, insert into the uterus. The ultrasonographer helped guide the physician in the positioning of the tip of the catheter to a suitable point near the cavity fundus, with care being taken to avoid contact with the fundal endometrium. The embryos were then slowly released at this point. After ET, the catheter was pulled back into the guide and taken out. Because the ET catheter is accommodated in the guide, it is possible to take it out without touching the cervix. The cells attached around the catheter were used in the study. The tip of the catheter was cut by about 3 cm and immersed in 350 μl of lysis buffer for RNA extraction. In order to confirm the cytological determination of endometrial cells or the expression of endometrial receptive markers, the tip of the catheter was cut by about 3 cm and immersed into the fixative.

Cell distribution and Immunocytochemistry
The endometrial cells were attached to a microscope slide (liquid-based cytology) and then stained with Papanicolaou stain (Papa) according to standard procedures. The cells were diagnosed as endometrial cells with a cytotechnologist (The International Academy of Cytology) (n=6). At the same time, the presence of cervical cells was confirmed. The cytokeratin (AE1/AE3) (n=3) and the expressions of putative endometrial receptive markers including estrogen receptor α (ERα, n=3), progesterone receptor (PR, n=3), and HOXA 10 (n=3) were evaluated by immunocytochemistry. One slide was prepared from one catheter, and each stain was performed on a different sample. AE1 and AE3 are specific for the acidic cytokeratin as well as the basic cytokeratin. The cocktail of AE1 and AE3 exhibits broad reactivity with two families of cytokeratin, acidic, and basic. The sections were treated in ethanol, and endogenous peroxidase was blocked with 3% H 2 O 2 . After a 60-min incubation with Protein Block Serum-Free (X0909 Dako, Santa Clara, USA), the sections were incubated for 60 min at room temperature with a primary antibody ( Table 1). The sections were then incubated with MAX-PO (MULTI). MAX-PO is a labeled polymer in which a peroxidase and a second antibody made into Fab are bonded to an amino acid polymer. Hematoxylin was used for counterstaining. Negative control sections were incubated without the primary antibody (n=3). Images were taken under a confocal microscope (Keyence BZ-9000, Osaka, Japan).

RNA Extraction and Quantitative Reverse Transcription-Polymerase Chain Reaction (RT-qPCR)
Total RNA was extracted from the cellular suspension using the RNeasy Micro Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions (n=60).
RT was performed using the iScript TM Advanced cDNA Synthesis Kit for RT-qPCR (BIO-RAD, Hercules, USA). All total RNAs were reverse transcribed in a 20-μl volume. The  Table 2). The expression of mRNA was normalized to RNA loading for each sample using GAPDH mRNA as an internal standard. These expression levels were calculated by the ΔΔCT method.

Statistical Analysis
Data are presented as the mean ± SD of duplicate samples and are presented as folds relative to the corresponding data, as the mean ± SD values were analyzed by Bonferroni-Dunn test. A level of p<0.05 was considered significant. The confidence intervals with p-values for multiple statistical analyses are at the 95% level.

Collection and Characterization of the Endometrial Cells
After ET, the tip of the catheter was cut by about 3 cm and immersed in the fixative. Cells attached to the catheter were evaluated for endometrial cells. The morphological evaluation by Papa staining suggested that endometrial epithelial cells were detected as the majority of cells and that they were not contaminated with cervical cells from the cervix neck ( Fig.  1a, b). These cells were confirmed and diagnosed as endometrial cells with a cytotechnologist. Additionally, immunocytochemical analysis revealed that positive staining for AE1/AE3 was detected in about 90% of the endometrial cells (Fig.  1c, d). Immunocytochemical assays were performed by at least two researchers in an operator-blinded manner.

Expression of Endometrial Receptive Markers
In order to investigate the protein expression in cells attached to the catheter, the expressions of ERα, PR, and HOXA 10 were confirmed and evaluated by immunocytochemistry in endometrial cells (Fig. 2).

mRNA Expression of Endometrial Receptive Markers
In order to investigate the mRNA expression in cells attached to the catheter, the three genes of mRNA expression were evaluated. When total RNA was extracted from the cellular suspension, the concentrations of extracted RNA were 3.7 to 43.4 ng/μl. Three genes (ERα, PR, HOXA 10) were evaluated that are associated with endometrial receptivity, and Fig. 3 shows the distribution of the three genes. The Ct values were converted to 2 −Ct . To examine the expression of the three genes by pregnancy result, our cohort was separated into two groups. The patient etiology is shown in Table 3. There was no difference between the pregnant group and nonpregnant group for age, blastocyst quality, endometrium thickness, concentration of estradiol, and concentration of progesterone at ET ( Table 4). The transferred blastocysts were scored and graded according to Gardner's criteria on day 5 [13]. Single embryo transfer was performed without genetic testing in all cycles. Recurrent implantation failure patients were excluded. When the expression in the pregnant group was 1, the expression in the non-pregnant group was 1.2 ±1.1-, 1.4±1.2-, and 2.8±1.5-fold, respectively, and the PR (p < 0.05) and HOXA10 (p < 0.01) mRNA expressions were significantly higher in the non-pregnant group (Fig. 4). ER α expression tended to be higher in the non-pregnant group, but there was no significant difference (p=0.18).

Discussion
Endometrial receptivity has been recognized as a potential source of implantation of an embryo. Histological evaluations have been regarded as the standard for the clinical diagnosis of endometrial abnormalities [14]. Recently, chronic endometritis due to common bacteria is reported in women with recurrent miscarriage [15][16][17]. Many studies have investigated the potential of proteomic analysis to characterize the expression patterns of physiologically active substances on proliferative and secretory endometrium [18][19][20]. The identification of gene expression profiles has led to a differential analysis of receptive and non-receptive patterns in endometrial signaling. The use of these expression patterns from endometrial tissues has been demonstrated to be an accurate and reproducible dating method to identify endometrial receptivity status. In the present study, our experimental results demonstrated that cells that came in contact with a catheter specifically positioned on the surface of the endometrium could provide insight into the level of endometrium receptivity. Indeed, using the cells attached to the catheter, we were able to evaluate the expression of genes associated with endometrial receptivity, and their expression differed by pregnancy outcome. It may be possible to identify the optimal luteal management or ETtiming by analyzing this difference in expression. Furthermore, not only gene expression but also protein expression could be confirmed by using this method. By profiling the transcriptome of 238 genes that are expressed at different stages of the endometrial cycle, the ERA was developed as a means of personalizing ET timing. It has been demonstrated to be reproducible in patients across multiple menstrual cycles and to be more accurate than histological analysis in defining the optimal window of implantation (WOI) [7]. Although this method has significant benefits, it does invade the endometrium, and it may not be able to reveal dell information related to the transfer cycle. That is, it is uncertain whether the endometrium of each cycle is the same status and whether there is any effect from biopsy. In addition, the cost of measurement is high. On the other hand, our method allowed for current cycle analysis, without an additional invasive procedure. Moreover, due to the nature of ET, this sample was taken without altering the ET. An important problem that we faced was that we can only collect Fig. 1 Endometrial cells attached to the catheter during embryo transfer. a, b Cells that attached to the catheter were stained by papanicolaou stain (n=6). Both cells were endometrial cells; b cells were simple columnar epithelium with pilus. c and d were stained by immunocytochemical staining of AE1/AE3, then visualized with a Keyence BZ-9000 microscope (n=3). Both cells were endometrial epithelium cells. (a, b, d ×400, c 200×) Fig. 2 Expression of putative endometrial receptive markers in the human luteal-phase endometrium. Cells that attached to the catheter were stained using immunocytochemistry (ERα, PR, HOXA10, negative control; n=3), then visualized with a Keyence BZ-9000 microscope (×400). ERα and PR were stained in both cytoplasm and nucleus. HOXA10 was stained only in the cytoplasm a small amount of cells, which limits the potential number of endpoints to investigate. In our case, when total RNA was extracted from the cellular suspension, the concentrations of extracted RNA were 3.7 to 43.4 ng/μl. It may be possible to analyze only 10 genes per sample by the usual PCR method. On the other hand, many gene analyze may be possible using arrays. In addition to this result, a comprehensive analysis will be performed to select the genes required for evaluation. The final aim of this study was to narrow down the gene profile at implantation that would have a greater potential of leading to pregnancy.
It was investigated whether ERα, PR, and HOXA 10 could be markers of endometrial receptivity. Although ERα and PR were not known as receptive markers, in this study, it was examined whether it could be a predictor or regulator of other genes. The expression of these genes has been reported to change during the menstrual cycle. During the transition from the early to mid-secretory phases of the menstrual cycle there is a cell-specific decrease in estrogen and progesterone receptors on the endometrial epithelium [21] Expression of HOXA 10 increased in the secretory phase [22].
These genes have been reported to be associated with implantation [23][24][25][26][27]. In our results, progesterone receptor expression was 1.4-fold and HOXA10 expression was 2.8fold significantly higher in patients who became nonpregnant group. Gene expression of HOXA 10 was reportedly is increased in a progesterone-dependent fashion in human endometrium [28]. On the other hand, gene expression of CK 15 is decreased in a HOXA10-dependent fashion in human endometrium epithelial cells [3]. It has been reported that there is a cell-specific decrease in estrogen and progesterone receptors in the endometrial epithelium during the transition from the early to a mid-menstrual cycle of the menstrual cycle. However, in our results, progesterone receptor expression was 1.4-fold significantly higher in patients who became a nonpregnant group. Therefore, high expression of progesterone receptors is thought to uptake progesterone, and as a result, induces high expression of HOXA 10. In the non-pregnancy cycle, there is concern that excessive HOXA10 expression will sharply reduce cytokeratin expression. Therefore, increasing HOXA10 may cause to be difficult to implantation by making cell structure fragile. In other reports, colonystimulating factor-1 and LIF expression in endometrial cells isolated from current cycle cannula cells showed associations with increasing endometrial receptivity and pregnancy [25]. Prior to that, it was reported that a significant proportion of patients with a history of failure of the implantation of a euploid embryo had a displaced WOI as detected by the ERA [29]. For these patients, personalized ET using a modified progesterone administration protocol may improve the outcomes of subsequent euploid ET. Our new method will be able to analyze the condition of the endometrium in order to optimize the luteal management or ET-timing. It is important to determine the gene profile at implantation that would have the greatest potential of leading to pregnancy.
This procedure that we have established made it possible to conduct analysis in the current cycle, without an additional invasive procedure. Moreover, the cells attached to the catheter were revealed information about the transfer site without altering the endometrium function, and with minimal contamination from other cells that came in contact with the catheter. In addition, this information may lead to ameliorate the endometrial  conditions with the next implantation, even if implantation of this cycle failed. Taken together, this procedure could prove to be a reliable method for detecting genes to assess endometrial receptivity. A limitation of our study is that we examined slight gene expressions in cells obtained from the surface of the endometrium. The remaining question is whether any other genes are present with modulating the implantation or the perturbation of gene expression could cause implantation failure in endometrium. In the future, including the relationship of blood adhesion to the catheter, concerning the comprehensive analysis between pregnant and non-pregnant will be needed.
In conclusion, our findings have revealed that our method can be used to take a sample of the endometrium, and this method does not affect the condition of ET nor is it invasive. This method can evaluate the state of the endometrium at the time of ET and may contribute to the improvement of implantation. Fig. 4 Transcriptional profile in endometrial cells with respect to pregnancy outcomes. The expression of ERα, PR, and HOXA10 was determined by RT-qPCR. The fold differences in expression levels were calculated according to the 2 -ΔΔCt method, in duplicates. The data are expressed as mean ± SD. Bonferroni-Dunn test determined PR (p<0.05) and HOXA10 (p<0.01) was significantly higher in nonpregnant endometrial cells than those who pregnant