Decreased CD44v3 Expression Impairs Endometrial Stromal Cell Decidualization in Women With Recurrent Implantation Failure

Background The precise pathogenesis of poor endometrial receptivity in recurrent implantation failure (RIF) still remains unclear. This study aims to explore the effects of different CD44 isoforms in the mid-secretory phase endometrium on endometrial receptivity in women with RIF. Methods Mid-secretory phase endometrial tissue samples were obtained from two groups of women who had undergone IVF: a) 24 patients with RIF, b) 18 patients with infertility due to tubal obstruction, who had achieved a successful clinical pregnancy after the rst embryo transfer in IVF (control group). Identication of differentially expressed CD44 isoforms in endometrial tissues was assessed with immunohistochemistry, qPCR and western blotting. Effects of CD44v3 overexpression and knockdown on proliferation and decidualization of Immortalized human endometrial stromal cells (T-HESCs) and primary HESCs were investigated by qPCR and Western blot. A heterologous co-culture system of embryo implantation was constructed to mimics the process of trophoblast invasion during implantation.


Introduction
Reproductive failure is a major social and economic problem. In recent years, breakthroughs in assisted reproductive technology have improved outcomes for couples who previously failed to conceive and achieve viable pregnancy 1 . Nevertheless, a new challenge has emerged, as approximately 10% of infertile women who undergo in vitro fertilization and embryo transfer (IVF-ET) cycles suffer from recurrent implantation failure (RIF) 2 . RIF refers to an unsuccessful clinical pregnancy after a minimum of three transfers of at least four morphologically good quality embryos ( grade above 3BB or score ≥ 7) into a normal uterus 3 . Disturbed immunological factors and maternal endocrine abnormalities, particularly inadequate endometrial receptivity, constitute some of the RIF etiologies 4 . However, the precise pathogenesis of poor endometrial receptivity in RIF still remains unclear.
During the secretory phase, the endometrium undergoes decidualization, and endometrial stromal cells proliferate and differentiate into decidual cells to attain endometrial receptivity 5 . Decidualization, which plays key roles in embryo support, nutritional furnishing, endocrine regulation, and immune modulation, is a prerequisite for embryo implantation in some mammals, including mice and humans 6 . For example, loss of BMP2 in the uterus renders mice infertile, causing failure of decidualization. Additionally, uterine Fst-cKO mice show severe fertility defects, including a poor decidualization response, with relatively low levels of stromal proliferation and differentiation 7 . Our previous study demonstrated that reduced PIBF1 expression in the mid-secretory phase of RIF patients inhibits HESCs proliferation and decidualization 8 .
The transmembrane glycoprotein CD44 is thought to participate in various cellular processes, including regulation of cell proliferation, migration, and adhesion 9 . Therefore, it may be necessary for pregnancy maintenance. For instance, the CD44-mediated inchoate attachment between endometrial epithelial cells and trophectoderm and relatively low CD44 expression in decidual cells were found to be associated with unexplained incidences of miscarriage 10,11 . Furthermore, it has been reported that CD44 is negatively expressed in human endometrial cells during the proliferative phase, whereas it shows intense expression in the mid and late secretory phases 12,13 . However, since CD44 frequently shows alternative spliced variants, including the shortest standard form of CD44 (CD44s) and multiple CD44 variants (CD44v), the unique isoforms of CD44 expressed in the endometrium during the menstrual cycle have not been fully elucidated. Thus, whether CD44 isoforms participate in regulating the development of endometrial receptivity in RIF is unclear. In this study, we explored the effects of different CD44 isoforms, particularly CD44v3, in the mid-secretory phase on endometrial receptivity in women with RIF.

Materials And Methods
Patients Participants were enrolled for 18 months from 2019 to 2021 at the Reproductive Medical Center of Ruijin Hospital. We recruited 24 RIF patients aged between 25 and 35 years who had experienced three or more previous failed cycles wherein at least four good quality embryos were transferred. The comparison group included 18 women with infertility due to tubal obstruction, who had achieved a successful clinical pregnancy after the rst embryo transfer in IVF. The exclusion criteria were as described previously 14 .
Brie y, individuals with uterus pathology, hydrosalpinx, adenomyosis, polycystic ovary syndrome, autoimmune disease, endometriosis and chromosome abnormalities were excluded. Endometrial samples were obtained through pipe suction curettage (LILYCLEANER; Shanghai Jiabao Medical Healthcare Science and Technology Ltd., China).
Furthermore, 12 women in the early proliferative (days 4-5 of the cycle) matching the same criteria as the control group were enrolled.

Cell culture
Immortalized human endometrial stromal cells (T-HESCs) and Ishikawa cells, common surrogates for human endometrial cells, were acquired from the European Collection of Authenticated Cell Cultures (ECACC; Salisbury, UK). Primary human endometrial epithelial cells (HEECs) and endometrial stromal cells (HESCs) were isolated as previously described. Brie y, endometrial samples were minced and digested for 30 min using 1 mg/mL collagenase type I (Thermo Fisher Scienti c, Waltham, MA, USA) at 37 °C. The mixture was then passed through 100 and 40 µm sieves successively (MilliporeSigma, Burlington, MA, USA), and the ushing and reverse ushing ltrates from the 40 µm sieve were centrifuged 5 min at 100 × g to isolate HESCs and HEECs,. T-HESCs, Ishikawa cells, primary HEECs, and HESCs were cultured in DMEM/F12 medium (Thermo Fisher Scienti c), supplemented with 10% (v/v) fetal bovine serum (FBS), 100 IU/mL penicillin, and 100 μg/mL streptomycin (Thermo Fisher Scienti c), at 37 °C according to standard procedures and harvested using 0.25% (w/v) trypsin-EDTA (Thermo Fisher Scienti c).

Protein isolation and western blotting
Proteins were extracted by lysing endometrial samples and cells with RIPA lysis buffer (Thermo Fisher Scienti c) supplemented with a protease inhibitor cocktail (Roche, Basel, Switzerland). The lysates were then centrifuged at 12,000 × g for 10 min at 4 °C, and the supernatant was collected. Samples with 30 μg proteins were separated via 10% SDS-PAGE, and the resolved proteins were transferred onto PVDF membranes (MilliporeSigma) that were blocked with 5% non-fat milk in TBST for an hour. The membranes were then incubated overnight at 4 °C with protein-speci c primary antibodies . Following washing and incubation with an corresponding HRP-conjugated antibody at room temperature for an hour, the bands were visualized via enhanced chemiluminescence (Millipore Sigma). Primary antibodies against CD44v3 (1 μg/mL; R&D Systems, Minneapolis, MN, USA) and GAPDH (1:1,000; Cell Signaling Technology, Danvers, MA, USA) were used in this study. Quanti cation was performed using the ImageJ software and normalized to GAPDH levels.

Immunohistochemical staining
Tissue specimens were xed with 4% formalin. Para n sections (5 μm) were prepared and xed. Antigen retrieval was performed by incubating the cells in buffered citrate for 15 min at 105 °C. The sections were blocked with 5% w/v bovine serum albumin for 30 min and then incubated with primary antibodies against CD44s (1:100; Abcam, Cambridge, UK), CD44v3 (10 μg/mL; R&D Systems), and CD44v6 (1:100; Abcam) overnight at 4 °C. The slides were then stained with horseradish peroxidase-conjugated secondary antibodies, followed by counterstaining with diaminobenzidine (Agilent Technologies, Santa Clara, CA, USA) and hematoxylin. Images were visualized using a microscope (Olympus Corporation, Tokyo, Japan).

Proliferation assay
Cell proliferation was examined using the Cell Counting Kit-8 (CCK-8; Dojindo China Co., Ltd., Shanghai, China). Brie y, cells were seeded in 96-well plates at a concentration of 1,000 cells per well. A total of 10 μL CCK-8 solution was added to each well, and the optical density was measured at a wavelength of 450 nm through a microplate reader. The experiment was repeated thrice.
In vitro decidualization activity assay HESCs separated from the late proliferative phase of the endometrim of the control group were transfected with CD44v3 siRNA or CD44v3 overexpression plasmids. After 48h transfection, the cells were cultured in a serum-free DMEM-F12 medium with 10 nM β estradiol (Sigma Aldrich), 1 μM progesterone (Sigma Aldrich), and 1 mM 8-Br-cAMP (Abcam) for 72 h. Total RNA and protein were extracted, the decidual markers prolactin (PRL) andinsulin like growth factor binding protein-1 (IGFBP1) were evaluated via RT-qPCR and western blotting.

Migration assay
Ishikawa cells (1 × 10 6 cells) with or without treatment were seeded in 6-well plates to reach sub con uence overnight. The monolayer cells were then scratched using a pipette tip to create a cell free wound. Cells were washed twice with PBS and then cultured in fresh serum-free medium. Wound healing ability was quanti ed by measuring the percentage of closure at 0, 24, and 48 h. Three independent experiments were conducted.

Embryo outgrowth analysis
Trophoblast outgrowth analysis was constructed as described. In Brief, HESCs were isolated from the late proliferative phase endometrium of the control group and then cultured in in a 24-well plate. After 48 h transfected with CD44v3 siRNA or CD44v3 overexpression plasmids, cells were decidualized as described above. Hatched mice blastocysts with normal morphology were then co-cultured with con uent monolayers of decidualized HESCs in DMEM/F12 complete medium. The trophoblast outgrowth areas were outlined and calculated using Image J 1.46r.

Statistical analysis
Data are presented as mean ± SEM and were analyzed using the SPSS software (version 22.0; SPSS Inc., Chicago, IL, USA). Statistical analysis between two groups was performed using the two tailed Student's t-test when data met the normal distribution criterion. For more than two groups, statistical analysis was performed using a one way analysis of variance (ANOVA) with the Bonferroni test for mean separation or a nonparametric test for non-normal data. Statistical signi cance was set at p < 0.05.

Demographics
Demographic details of the women recruited in the study are summarized in Table 1. Except for the number of embryos transferred, other indices were not signi cantly different between the control and RIF groups (p > 0.05).
Identi cation of differentially expressed CD44 isoforms in the RIF and control groups In the endometrial tissues from the mid-secretory phase, using RT-qPCR, no difference was found in the mRNA levels of CD44s, CD44v3, or CD44v6 between the control (n = 18) and RIF groups (n = 24) (Fig. 1A, p < 0.05). Immunostaining was then performed to investigate the cellular localization and protein levels of CD44 isoforms in human endometrial tissue. The results showed that CD44s and CD44v3 were localized in the epithelial and stromal cells of the endometrium (Fig. 1B). CD44v6 had three different expression patterns (Fig. 1B, Supplemental gure 1 CD44s and CD44v6 showed similar expression in both groups; however, the protein levels of CD44v3 were signi cantly reduced in the RIF group compared with those in the control group (Fig. 1B-E, p < 0.05), particularly in stromal cells (Fig. 1B, p < 0.01). To evaluate the clinical signi cance of this nding, a receiver operating characteristic (ROC) curve was derived to measure the performance of a binary classi er. As shown in Fig. 1F, the ROC area under the curve was 0.75 for CD44v3 tissue measurements, indicating a potential functional role of CD44v3 in the diagnosis of endometrial receptivity.

Characterization of CD44v3 expression in endometrial tissues and primary endometrial cells after E 2 and P 4 treatments
As assessed from the immunostaining assay results, CD44v3 expression was signi cantly higher during the mid-secretory phase (n = 12) than during the early proliferative phase of the endometrium (n = 15), especially in the endometrial stromal cells ( Fig. 2A-D, p < 0.05). Furthermore, we found that when stimulated with E 2 and/or P 4 , CD44v3 levels in primary endometrial stromal cells were signi cantly increased, with no signi cant change in primary endometrial epithelial cells (Fig. 2E-H, p < 0.05).

Effects of CD44v3 knockdown/overexpression on proliferation and decidualization of T-HESCs
The effects of CD44v3 knockdown and overexpression in T-HESCs were examined using western blotting ( Fig. 3A-D, p < 0.01). Knockdown or overexpression of CD44v3 did not affect the migration rate of T-HESCs ( Fig. 3E-F). Notably, cell proliferation decreased following CD44v3 knockdown and increased following CD44v3 overexpression (Fig. 3G, p < 0.05). To explore whether CD44v3 is involved in the decidualization pathways, an in vitro decidualization model of T-HESCs was constructed by treating the cells with E 2 , P 4 , and 8-bromo-cAMP. The results showed that mRNA expression levels of the two decidualization markers, PRL and IGFBP1, decreased following CD44v3 knockdown in T-HESCs. However, the levels of IGFBP1 increased after CD44v3 overexpression (Fig. 3H, p < 0.05).

Effects of CD44v3 knockdown/overexpression on proliferation and decidualization of primary HESCs
Similarly, the effects of CD44v3 knockdown and overexpression in primary HESCs were examined using western blotting (Fig. 4A-D, p < 0.05). CD44v3 knockdown/overexpression did not affect cell migration rate (Fig. 4E-F). Cell proliferation (Fig. 4G) and the expression of PRL and IGFBP1 (Fig. 4H-K, p < 0.05) were signi cantly reduced following CD44v3 knockdown. In addition, cell proliferation (Fig. 4G, p < 0.05) and the expression of PRL and IGFBP1 (Fig. 4H-K, p < 0.05) were signi cantly increased following CD44v3 overexpression. We further extended our observations to a heterologous co-culture system of embryo implantation, which mimics the process of trophoblast invasion during implantation in vivo. After 24 h of co-culture, the CD44v3-knockdown HESCs displayed a marked de ciency in supporting trophoblast outgrowth, whereas CD44v3-overexpressing HESCs signi cantly promoted trophoblast outgrowth (Fig. 5, p < 0.05). Data are presented as mean ± SD. The difference between two groups was analyzed by an independent sample t-test, except for the number of embryo transfers calculated using the independent samples Mann-Whitney U test (median, range). BMI, body mass index; FSH, follicle stimulating hormone; LH, luteinizing hormone; E 2 , estradiol.

Discussion
RIF with impaired endometrial receptivity remains a signi cant challenge in assisted reproductive technology. The rates of implantation, even for perfectly healthy blastocysts, are still dependent on adequate decidualization of the uterus, allowing it to be receptive to the embryo 15,16 . In the present study, we demonstrated, for the rst time, that the expression levels of CD44v3 were decreased in the midsecretory phase of the endometrium in women with RIF, which affected endometrial receptivity via the inhibition of endometrial stromal cell proliferation and decidualization.
CD44 family members are widely expressed transmembrane glycoproteins that participate in various cellular processes, including regulation of cell proliferation, division, migration, and adhesion 9 . Moreover, CD44 glycoproteins are critical mediators of tumorigenesis, endometriosis, and embryo epithelial interaction 17 . However, the functional role of CD44 family members in regulating endometrial receptivity in RIF patients remains unknown. During the mid-secretory phase, when embryo implantation occurs, we found that both CD44s and CD44v3 were widely expressed in glandular epithelial, luminal epithelial, and stromal cells. Although CD44v6 was previously reported to be restricted to the epithelial glands during the secretory phase, our results showed that CD44v6 was expressed in both the endometrial epithelium and stroma during the mid-secretory phase.
We then explored expression of the CD44 family members in the RIF and control groups. No differences were found in endometrial CD44s and CD44v6 expression between the two groups. For the rst time, the present study con rmed that CD44v3 expression signi cantly decreased in the mid-secretory endometrial stromal cells of patients with RIF than in those with secondary infertility, but with no statistically signi cant changes in endometrial epithelial cells. Furthermore, ROC analysis revealed high sensitivity and speci city to discriminate implantation status based on the expression levels of CD44v3 in endometrial tissues. To further investigate the relationship between CD44v3 and endometrial receptivity, we performed immunohistochemical analysis and western blotting to determine CD44v3 expression during the menstrual cycle in normal women. CD44v3 was found to be expressed at signi cantly higher levels during the mid-secretory phase of human endometrial stromal cells than during the proliferation phase. This study also aimed to identify the possible inducers of the upregulated CD44v3 expression during the mid-secretory phase. During the menstrual cycle, ovarian steroid hormones interact to prepare the endometrium for implantation. Accordingly, estradiol has been reported to increase CD44 expression 18,19 . Our results showed that CD44v3 was induced by E 2 and P 4 in primary HESCs.
Collectively, these ndings suggest that the decreased CD44v3 expression in the mid-secretory endometrial stromal cells may be a vital factor in RIF.
It has been established that endometrial stromal decidualization is crucial for successful embryo implantation. To create a receptive environment for blastocysts, endometrial stromal cells undergo proliferation during the estrogen dependent phase of the menstrual cycle, followed by differentiation into decidual secretory cells under the in uence of progesterone and estrogen 7,20 . CD44v3 (CD44v3-v10), containing CD44 variable exons 3-10, has been con rmed to play an important role in proliferation in various types of cancer cells 21 . In the present study, we found that cell proliferation was decreased following CD44v3 knockdown, and the proliferation rate increased following CD44v3 overexpression in both T-HESCs and HESCs, consistent with the results of a recent study demonstrating that CD44 promotes decidual stromal cell (DSC) proliferation and growth by binding to high molecular weight HA 11 .
Since the appropriate proliferation of endometrial stromal cells is necessary for decidualization, we further investigated the effect of CD44v3 on stromal cell decidualization. Through an in vitro decidualization assay, we found that the expression of PRL and IGFBP1, two decidualization markers, was notably decreased following CD44v3 knockdown and signi cantly increased following CD44v3 overexpression in HESCs. Furthermore, the CD44v3-knockdown HESCs displayed a marked de ciency in supporting trophoblast outgrowth, whereas the CD44v3-overexpressing HESCs promoted trophoblast outgrowth. In Ishikawa cells, there was not much difference in cell migration, proliferation, and adhesiveness capabilities after CD44v3 overexpression/knockdown (Supplemental gure 2). Overall, we showed that the decreased CD44v3 expression inhibited the proliferation and decidualization of stromal cells in women with RIF, which might contribute to impaired endometrial receptivity.

Conclusions
We demonstrated for the rst time that CD44v3 expression was higher during the mid-secretory phase of human endometrial stromal cells than during the proliferation phase, with a signi cant decline in the endometrial stroma of women with RIF. This indicates that low CD44v3 expression might suppress stromal cell proliferation and decidualization, further impairing endometrial receptivity in women with RIF.

Declarations
Ethics approval and consent to participate The protocol for this study was approved by the ethical committee of Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University (2012-57). All participants provided written informed consent.

Consent for publication
Not applicable.

Availability of data and materials
The datasets in this study are available from the corresponding authors on reasonable request.

Competing interests
The authors declared that they have no competing interests.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.