Human chorionic gonadotropin promotes recruitment of regulatory T cells in endometrium by inducing chemokine CCL2

Journal Pre-proof Human chorionic gonadotropin promotes recruitment of regulatory T cells in endometrium by inducing chemokine CCL2 Xiaomin Huang, Yunni Cai, Min Ding, Bo Zheng, Haixiang Sun, Jianjun Zhou

PII:

S0165-0378(19)30434-6

DOI:

https://doi.org/10.1016/j.jri.2019.102856

Reference:

JRI 102856

To appear in:

Journal of Reproductive Immunology

Received Date:

28 July 2019

Revised Date:

8 October 2019

Accepted Date:

14 November 2019

Please cite this article as: Huang X, Cai Y, Ding M, Zheng B, Sun H, Zhou J, Human chorionic gonadotropin promotes recruitment of regulatory T cells in endometrium by inducing chemokine CCL2, Journal of Reproductive Immunology (2019), doi: https://doi.org/10.1016/j.jri.2019.102856

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier.

Human chorionic gonadotropin promotes recruitment of regulatory T cells in endometrium by inducing chemokine CCL2 Xiaomin Huang, Yunni Cai, Min Ding, Bo Zheng, Haixiang Sun*, Jianjun Zhou *

Reproductive Medicine Center, Department of Obstetrics and Gynecology, The

China

Address all correspondence and requests for reprints to:

ro of

Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing,

-p

Jianjun Zhou, Reproductive Medicine Center, Department of Obstetrics and

Gynecology, The Affiliated Drum Tower Hospital of Nanjing University Medical

70014,

lP

E-mail: [email protected]

re

School, 321# Zhongshan Road, Nanjing 210008, P.R.China. Tel/fex:+86258310666-

na

Highlights 

Intrauterine infusion of hCG increased the endometrial Tregs and CCL2 expression



CCL2 expression and Tregs migration were significantly increased with hCG

ur

stimulation

Tregs migration was decreased with CCL2 siRNA or CCR2 antagonist treatment



Intrauterine infusion hCG promotes the recruitment of Tregs by inducing CCL2

Jo



Abstract

Human chorionic gonadotropin (hCG) can attract regulatory T cells (Tregs) into the fetal-maternal interface regulating maternal immune tolerance in pregnancy. The objective of this study was to investigate whether hCG recruits the Tregs into endometrium by inducing chemokines. The number of Tregs in the endometrium was

analyzed by immunohistochemistry. The expression of CCL2 was analyzed in vivo and in vitro with hCG stimulation. CD4+CD25+ Tregs were isolated from peripheral blood for Tregs migration assay with hCG, CCL2 siRNA and CCR2 antagonist stimulation. Our results showed that the number of endometrial Tregs in RIF patients was significantly decreased (9.4±5.3 vs. 23.1±3.1, P<0.01), while intrauterine infusion of 2000 IU hCG increased the endometrial Tregs (18.6±9.8 vs. 9.4±5.3, P<0.05) and CCL2 expression (0.21±0.01 vs. 0.17±0.01, P<0.01). Horn injecting with 10 IU hCG also increased the endometrial Tregs in pseudopregnant mice (46±16.8 vs.

ro of

7±4.3, P<0.01). Furthermore, the CCL2 protein and mRNA levels were significantly increased in human endometrial stromal cells (hESCs) with the stimulation of hCG.

Migration assays showed that hESCs with hCG stimulation promoted Tregs migration (2597±833.2 vs. 1115±670.7, P<0.05), while the number of migrated Tregs

-p

significantly decreased with CCL2 siRNA (1061±105.4 vs. 2598±294.7, P<0.05) or CCR2 antagonist (356.7±138.8 vs. 2597±833.2, P<0.05) treatment. In conclusion,

re

intrauterine perfusion of hCG might promote the recruitment of Tregs into

lP

endometrium by inducing chemokine CCL2.

Keywords: hCG, CCL2, Treg, maternal-fetal immune, RIF

na

1. Introduction

Repeated implantation failure (RIF) was defined as patients who are under the age of

ur

40 and had ≥ 3 consecutive transfers of ≥ 4 high-grade embryos while have failed pregnancy (Coughlan et al., 2014), about 10% infertility patients suffer from RIF who

Jo

underwent in vitro fertilization (IVF) treatment. Embryo transplantation is a unique immunological challenge as embryo is semi-antigen to maternal immune system, so immunological factor is thought to play an important role in the pathophysiological mechanism of RIF (Bashiri et al., 2018). Regulatory T cells (Tregs) have been described as a specific subpopulation of T lymphocytes, which play a major role in transplantation tolerance and pregnancy establishment (Ghaebi et al., 2017). Our previous study also showed that increased Tregs was associated with increased rates

of pregnancy and live birth in IVF treatment (Zhou et al., 2012), while some studies and our results showed that the endometrial Tregs in women with RIF were significantly lower than that in the control women (Diao LH et al.,2017; Jiang et al., 2017). So Treg cells might be the therapeutic target of RIF. Human chorionic gonadotropin (hCG) is a glycoprotein hormone, which is produced by placental syncytiotrophoblast cells during pregnancy, and is a powerful paracrine regulator induces successful implantation and coordinated dialog between blastocyst and endometrium (Chinedu et al., 2017). A previous study showed that

ro of

hCG can attract regulatory T cells into the fetal-maternal interface during early human pregnancy (Schumacher et al., 2009). So intrauterine infusion of hCG before embryo

transfer was used to promote embryo implantation (Fatemeh et al., 2017; Zhang et al., 2019). A meta-analysis of randomized controlled trials showed that intrauterine

-p

injection of hCG before embryo transfer can improve IVF-ET outcomes (Gao M et al., 2019), and another meta-analysis found that intrauterine injection of hCG only

re

worked when using of hCG ≥ 500 IU (Craciunas L et al., 2018). However, there were also some studies showed that hCG had no effect on the live birth rate (Osman A et

lP

al., 2016). Therefore, we hypothesized that hCG had a role in improving pregnancy outcomes in some specific populations, such as RIF patients with decreased Tregs. A

na

prospective cohort study showed that intrauterine administration of hCG improved the live birth rate of RIF patients (Liu X et al., 2019) and a retrospective study found that intrauterine administration of hCG before FET significantly improved the pregnancy

ur

rates, especially after one and three or more implantation failures (Huang P et al., 2018). However, the mechanism of hCG intrauterine infusion promote embyro

Jo

transplantation is unclear.

Studies showd that hCG can regulate the expression of chemokines in

endometrial cells (Borisova et al., 2017; Paiva et al., 2011 ). A liquid suspension chip detection result implied that the expression of chemokines was increased in human endometrial cells stimulated by hCG (Srivastava et al., 2013). So hCG might be a chemoattractant for Tregs to supports maternal-fetal immune tolerance (Borisova et al., 2017). The objective of this study was to investigate whether hCG recruits the

Tregs into endometrium by inducing chemokines.

2. Materials and methods 2.1 Subjects A total of nine women with RIF and ten control women were enrolled for this study from June 2018 to January 2019. Women with RIF inclusion criteria: failure of clinical pregnancy after ≥ 4 good quality embryo transfers, with ≥ 3 embryo transfer cycles and age < 40. Control women inclusion criteria: patients under the age of 40

ro of

underwent their first IVF-ET due to fallopian tube factors and got live birth at the first embryo transfer. Exclusion criteria: patients who had endometritis, endometriosis,

hysteromyoma, adenomyosis or uterine malformation, patients who had hereditary thrombophilia, abnormal auto-antibodies or antiphospholipid syndrome. The

-p

endometrial biopsies of participants were obtained at implantation window (LH+7-9d of the menstrual cycle) and the expression of Foxp3 (clone SP97, 1:400; Abcam,

re

Cambridge, UK) was analyzed by immunohistochemistry. And nine women with RIF were recruited to perform intrauterine infusion of 2000IU hCG at LH+5-7d of the

lP

menstrual cycle and 48 hours after intrauterine infusion endometrial biopsies were obtained to analyze the expression of Foxp3 and CCL2 (25542-1-AP, 1:400;

na

Proteintech, Chicago, USA). All samples were collected with the informed consent of the patients, and approval from the ethics committee in this study. .

ur

2.2. Immunohistochemistry (IHC)

Endometrial samples were routinely fixed and embedded in paraffin wax. After

Jo

antigen retrieval and blocking for 30 minutes, slides were incubated with rabbit antihuman antibody against Foxp3 and CCL2 overnight at 4°C. A negative control was stained by an equivalent concentration of rabbit IgG and did not detect any specific staining. The numbers of Foxp3+ cells were counted visually in five random microscopic fields at 200× magnifications for each biopsy by two blinded people. The integrated optical density (IOD) to area ratios of CCL2 were tested in five random fields of view

by Image pro-plus software (Media Cybernetic, Rockville, USA). Digital images were taken by using a Leica DM 2000 microscope and LAS Core software (Leica Microsystems Limited, Wetzlar, Germany). . 2.3. Mice model of horn injecting with hCG As the stromal cells in mice endometrium also express luteinizing hormone receptor (LHR) (Zheng et al., 2001; Lin et al., 2005), mice model of horn injecting with hCG was also used. ICR mice (6–8 weeks old) were obtained from the Nanjing Medical

ro of

University Animal Center (Jiangsu Province, China). We applied pseudopregnant mice which were mated with proven ligation males. At 1.5 days post-coitus (dpc),

pseudopregnant female mice were injected of 5 IU hCG into each uterine horn. And the control groups were injected of equal volume PBS. At 3.5 dpc, five mice per

-p

group were euthanized by cervical dislocation and uteruses were obtained for IHC

routinely. The expression of Foxp3 (D6O8R, 1:400; CST, Danvers, MA, USA) was

re

analyzed by IHC and counted with the method mentioned before.

lP

2.4. Isolation of hESCs and stimulation in vitro

Human endometrial stromal cells (hESCs) were isolated from mid-secretory phase

na

endometrial tissues collected from patients in our center. These cells were isolated and cultured as previously described (Liu et al., 2013). After starved in medium without fetal bovine serum overnight, the hESCs were cultured in DMEM/F12 medium

ur

(HyClone, CORNING, South Logan, UT, USA) containing 10% (v/v) charcoal/dextran-treated fetal bovine serum (FBS; Gibco), 100 IU/ml penicillin, and

Jo

100 μg/ml streptomycin supplemented with 0, 20, 40, 100 IU/ml of hCG for the different durations as 0, 4, 12, 24, 48 hours.

2.5. Isolation of Tregs For migration assays, CD4+CD25+ Tregs were isolated from peripheral blood samples from the control women. The Tregs were obtained using magnetic beads from Miltenyi Biotec, following the instructions of the manufacturer. After isolation, Tregs

were counted and floated in Opti-MEM medium and prepared for migration assays. The purity of the isolated Tregs was determined before using the cells for migration assays by staining with CD4-FITC (eBioscience, Colorado, USA) and CD25-APC (eBioscience) and analyzing them by flow cytometry. The purity of the isolated cells varied between 92 and 97% (Fig 5b).

2.6. Tregs migration assays A two-chamber system was established to conduct Tregs migration assays. hESCs

ro of

were put in the lower chamber and Tregs were put in the upper chamber. hESCs were trypsinized with 0.05% trypsin/EDTA (Gibco) and plated overnight at 1×105

cells/well in 24-well plates. For migration assays, the medium was changed to Opti-

MEM medium (1.3 mL/well). Cell inserts (8 μm, Milicell, CORNING) were placed in

-p

the wells. Then, 200 μL of Opti-MEM medium containing isolated Treg (2×104/well) was filled in each insert. After 24h, inserts were removed and supernatants from the

re

lower chamber were taken. Cells were washed with PBS and incubated for 30 minutes with CD4-FITC (eBioscience) and CD25-APC (eBioscience). After washing, the

lP

absolute number of migrated Tregs was determined by flow cytometry (FACScan; Becton Dickinson, USA). Tregs migration assay were performed with hCG, CCL2

na

siRNA treatment of hESCs and CCR2 antagonist RS 102985 hydrochloride (Sigma) stimulation with Treg cells.

ur

2.7. Immunofluorescence staining for LHR hESCs grown in 24-well chambers (Millipore, CORNING, MA, USA) were fixed

Jo

with 4% paraformaldehyde (w/v) for 30 min at room temperature. Next, the cells were washed with PBS and permeabilized with 0.5% Triton X-100 in PBS at room temperature. Subsequently, the cells were blocked with 3% BSA in PBS at 37 °C for 45 min and incubated with LHR (bs-6431R, 1:400; Bioss, Beijing, China) at 4 °C overnight. Finally, the cells were visualized using a fluorescence microscope (Leica, Wetzlar, Germany).

2.8. Western Blot The cells at passage 4 were gathered to extract protein for next experiments by routine methods. The concentrations of protein were determined by the BCA Protein Assay Reagent (Thermo Fisher Scientific, Rockford, IL, USA), and 10 μg of total proteins was loaded to 10% SDS-PAGE gel electrophoresis and transferred to a PVDF membrane (PVDF, Millipore). The membrane was incubated with primary antibodies against CCL2 (1: 4000, ab151538, Abcam) or GAPDH (1: 10000, AP0063, Bioworld technology), followed by incubation with goat anti-rabbit (1: 10000, A0545, Sigma).

ro of

The HRP activity was detected by an enhanced chemiluminescence kit (Amersham Biosciences Corp., Piscataway, NJ, USA), and densitometric analysis of each band

2.9. Quantitative Real-Time PCR (qPCR)

-p

was performed with UN-SCAN-IT (Silk Scientific, USA) software.

Total RNA was extracted from hESCs using Trizol reagent (Vazyme, Nanjing, China),

re

according to the manufacturer’s instructions, and was reverse transcribed in cDNA by using 5X All-In-One RT Master Mix (abm, Richmond, BC, Canada). The mRNA

lP

level was measured with the SYBR Green master mix (Vazyme) and the Roche LightCycler96 fluorescent quantitative PCR (LightCycler96, Roche, Basel, Sweden)

na

according to the manufacturer’s instructions. The primers of the indicated genes were listed (Table 1), and h18S rRNA was used as control. Reactions were run in duplicate by three independent experiments. The fold change in expression of each gene was

ur

calculated using the 2-△△CT method.

Jo

2.10. Statistical analysis

The values were presented as the mean and standard deviation. For the in vitro experiments, between- group differences were compared using Student’s t test or oneway ANOVA with Bonferron correction.

3. Results 3.1. Endometrial Tregs decreased in women with RIF

There was no significant difference in age, basal FSH, basal LH or antral follicle count between the RIF group and the control group, while the times of embryo transfer (3.9±1.1 vs. 1.0±0.0, P=0.001) and the number of transferred embryos (6.3±2.0 vs. 1.7±0.5, P=0.002) were increased in women with RIF compared with the control women (Table 2). Immunohistochemistry assays showed that Foxp3+ Treg cells were mostly confined to the stroma in the mid-secretory endometrial tissue, and the number of endometrial Foxp3+ cells in women with RIF was significantly lower

ro of

than that in the controls (9.4±5.3 vs. 23.1±3.1, P<0.01) (Fig 1).

3.2. Intrauterine perfusion of hCG increased endometrial Tregs

The number of endometrial Foxp3+ Tregs after intrauterine infusion of hCG in women

(18.6±9.8 vs. 9.4±5.3, P<0.05) (Fig 2a and b).

-p

with RIF was significantly higher than that before intrauterine infusion of hCG

Although the pregnant mice do not produce hCG, but the stromal cells in mice

re

endometrium still express LHR, so we established a pseudopregnant mice model that injected 5IU of hCG into each horn at 1.5 dpc and obtained the uterus after 48 hours.

lP

The number of endometrial Foxp3+ cells in pseudopregnant mice whose horn injected hCG were significantly increased than that in the injected PBS group and the no

na

surgery group (46±16.8 vs. 7±4.3, P<0.01, 46±16.8 vs. 6.2±0.6, P<0.01, respectively) (Fig 2d, e and f), while the injected PBS group and the no surgery group had no

ur

difference (7±4.3 vs. 6.2±0.6, P>0.05) (Fig 2e and d).

3.3. Chemokine CCL2 increased with hCG stimulation in vivo and in vitro

Jo

Immunofluorescence demonstrated that LH/CG receptors were expressed in the cytoplasm and nucleus of endometrial stromal cells (hESCs) (Fig 3). We then study the changes of chemokines CCL2, CCL3, CCL4, CCL5, CCL7 and CXCL10 mRNA levels in hESCs with hCG stimulation. Our study showed that only the chemokines CCL2 was significantly increased with hCG stimulation (7.10±0.72 vs. 1.00±0.72, P<0.05), while CCL3, CCL4, CCL5, CCL7 or CXCL10 had no significant difference (Figure 4a). Meanwhile, the concentration of CCL2 in the supernatant of hESCs with

hCG stimulated was significantly higher than that without hCG (256.5±51.15 vs. 201.4±39.39 pg/mL, P<0.05) (Figure 4b). Furthermore, we used Western Blot to detect changes in the expression of CCL2 protein with different concentration and time of hCG stimulation. We found that the expression of CCL2 was highest at 100IU/mL hCG and stimulation of 24 hours (Fig 4c-f), so the Tregs migration assays were carried out in accordance with this concentration and stimulating time. The results of IHC showed that the expression of CCL2 in the control group was

ro of

significantly higher than that in the RIF patients before intrauterine perfusion of hCG (0.19±0.004 vs. 0.17±0.01, P<0.01). In women with RIF, the results of IHC showed

that the CCL2 protein level in endometrium after intrauterine perfusion of hCG was higher than that before intrauterine perfusion of hCG (0.21±0.01 vs. 0.17±0.01,

-p

P<0.01) (Fig 4g-j). In conclusion, the expression of CCL2 was increased with

re

stimulation of hCG in vitro and in vivo.

3.4. hCG promoted the recruitment of Tregs toward hESCs by enhancing the

lP

expression of the chemokine CCL2

A two-chamber transwell system was established to analyze the number of migrated

na

Tregs toward hESCs with hCG, CCL2 siRNA and CCR2 antagonist stimulation (Fig 5a). Flow cytometry showed that hCG stimulation of hESCs could increase the number of migrated Tregs toward hESCs (2597±833.2 vs. 1115±670.7, P<0.05) (Fig

ur

5j and h). Furthermore, we proceeded to explore the effect of CCL2 on Tregs migration by down-regulating the expression of CCL2 with siRNA and adding CCR2

Jo

antagonist. Flow cytometry showed that the number of migrated Tregs was decreased when the expression of CCL2 in hESCs was down-regulated by siRNA (1061±105.4 vs. 2598±294.7, P<0.05) (Fig 5f and e, Fig 6). Meanwhile, the number of migrated Tregs was also decreased when the receptor of CCL2 was blocked in Tregs by CCR2 antagonist (356.7±138.8 vs. 2597±833.2, P<0.05) (Fig 5k and j).

4. Discussion

In this study, we found that intrauterine infusion of hCG increased the endometrial Tregs and CCL2 expresssion in RIF patients, while the Tregs migration was blocked with CCL2 siRNA or CCR2 antagonist treatment in vitro, so intrauterine perfusion of hCG might increase the endometrial Tregs recruitment by inducing chemokine CCL2. Tregs are indispensable important immune cells that regulating maternal immune tolerance in pregnancy. There were several researches showed that endometrial Tregs began to increase during the peri-implantation period，and reached the highest level at 2nd trimester (Pijnenborg et al., 1983). Tregs have been demonstrated that they

ro of

regulated immune reaction for embryo implantation by releasing IL-10, TGF-β and sequestering IL-2 (Takenaka et al., 2017; Schumacher et al., 2012). However, women with reproductive disturbances, such as spontaneous abortion and RIF, had less endometrial Tregs than normal pregnant women (Ren et al., 2019). Besides,

-p

endometrial Tregs decreased in women with RIF in this study, which was consistent with results of others (Jiang et al., 2017; Ghaebi et al., 2019). The pregnancy rate is

re

increased when the number of Tregs in patients with RIF is increased. A clinical research showed that the pregnancy rate was increased in the treated group in

lP

comparison with the control group, and the number of Tregs was also increased in the treated group (Ahmadi et al., 2017). A study also showed that the level of Tregs was

na

increased after the therapy of hCG plus immunoglobulin in patients with unexplained recurrent spontaneous abortion (Sha et al., 2017). Also, in this study, intrauterine perfusion of hCG in RIF patients could increase the endometrial Tregs and horn

ur

injected with hCG in pseudopregnancy mice model also could increase the endometrial Tregs. These results implied that the number of Tregs would alter in

Jo

reproductive disturbances and Tregs played very important role in embryo implantation.

The mechanism of Tregs mediating implantation success is complex. Originally,

Tregs in paraaortic lymph nodes (PALNs) were produced with the stimulation of semen antigen and released into the peripheral blood (Guerin et al., 2011). In the PALNs, semen antigen was presented to naive Th0 cells, and naive Th0 cells were activated, proliferate, and differentiate into Tregs to release into the peripheral blood

(Aluvihare et al., 2004). At the same time, peripheral Tregs (pTregs) and thymusderived Tregs (tTregs) migrated into endometrium, and the number of endometrial Tregs increased (Ghaebi, M., et al., 2017). In early pregnancy, decidual stromal cells expressed a large amount of chemokines and established the chemokine network at the maternal–fetal interface in order to recruit various immune cells such as natural killer (NK) cells, T cells and macrophages (Ramhorst et al., 2016). The expression of CCL2, CCL22, CCL17, CCL4 and CCL20 in hESCs were all increased in early pregnancy and the receptors of these chemokines were found expressed on the surface

ro of

of Tregs (Du et al., 2014). It has also reported that seminal fluid induced expression of chemokine CCL19 in glandular and luminal epithelial cells, which could act with the CCR7 receptor on the surface of Tregs (Guerin et al., 2011). Therefore, it is

speculated that pTregs might be migrated into endometrium by these chemokines.

-p

However, the mechanism of Tregs migrating into endometrium is still unclear.

Previous study showed that LH/CG receptors were expressed in the human

re

uterus, as well as in endometrial stromal cells (Reshef et al., 1990). We also demonstrated that LH/CG receptors were expressed in the cytoplasm and nucleus of

lP

endometrial stromal cells by immunofluorescence. This indicated that hCG regulated the function of endometrial stromal cells by LH/CG receptors. The importance of

na

hCG in regulating maternal fetal immunity had been confirmed (Bansal et al., 2012). It had been proven that hCG could increase the number of migrated Tregs in vitro (Schumacher et al., 2009). However, the mechanism is still unclear. It had been known

ur

that monocyte chemoattractant protein-1 (MCP-1/CCL2) was significantly elevated in decidual stromal cells in early pregnancy and was regulated by hCG (He et al., 2007).

Jo

Also, two reported proteomic analyses showed that a variety of chemokines including CCL2 would increase in the endometrial cells after stimulation of hCG (Paiva et al., 2011; Srivastava et al., 2013). Similarly, we examined chemokines CCL2, CCL3, CCL4, CCL5, CCL7 and CXCL10 mRNA levels in hESCs after hCG stimulation, and found that only the CCL2 was significantly elevated. Furthermore, western bolt demonstrated that the CCL2 protein level was also increased in hESCs after hCG stimulation in this study. The expression of CCL2 in women with RIF was found

significantly lower than the control women, but increased after intrauterine perfusion of hCG. hCG had been proven that could increase the number of Tregs in endometrium. However, the mechanism of hCG increasing the recruitment of Tregs into endometrium still need to study. And in early pregnancy, the endometrium secretes a large number of chemokines that mediate the migration of immune cells to the endometrium to regulate the maternal immune (Du et al., 2014). CCL2 is one of the key chemokines that regulate the recruitment and activation of lymphocytes, which

ro of

was also known to be associated with Tregs migration (Zhang et al., 2009). CCL2 have been proven to migrating Tregs in rheumatic immune disease and aggressive

cancers. In rheumatoid arthritis patients, the number of Tregs increased with elevated expression of CCL2 (Vasilev et al., 2019). In glioblastoma multiforme, elevated

-p

expression of CCL2 promoted recruitment of Tregs to help tumors evade immune

responses in tumor microenvironment and reduced overall survival of patients (Panek

re

et al., 2019; Chang et al., 2016). Although these studies have demonstrated the relationship between CCL2 and Tregs, the role of CCL2 in the endometrium remains

lP

unclear. Meanwhile, Tregs were proved can express the CCL2 receptor CCR2 (Du et al., 2014). Our study demonstrated hESCs with hCG stimulation promoted Tregs

na

migration, while the number of migrated Tregs was decreased when the expression of CCL2 in hESCs was down-regulated by siRNA. Meanwhile, the number of migrated Tregs was also decreased when the receptor of CCL2 in Tregs was blocked by CCR2

ur

antagonist. These results demonstrated that hCG promoted CCL2 expression in endometrial stromal cells recruiting Tregs into endometrium. However, the

Jo

mechanism hCG regulates CCL2 expression is still unclear. In conclusion, this study demonstrated that hCG induced the recruitment of Tregs

into endometrium by stimulating secretion of chemokine CCL2 expression. Intrauterine infusion of hCG might be a new therapy for Tregs-decreased RIF patients, which need to be explored in a larger prospective study.

Declarations of interest: none.

Acknowledgments This study was supported by National Natural Science Foundation of China (81571504, 81771537) and Natural Science Foundation of Jiangsu Province (Grants

Jo

ur

na

lP

re

-p

ro of

No BK20181117).

References Ahmadi, M., et al., 2017. Regulatory T cells improve pregnancy rate in RIF patients after additional IVIG treatment. Syst Biol Reprod Med. 63, 350-359. Aluvihare, V. R., et al., 2004. Regulatory T cells mediate maternal tolerance to the fetus. Nat Immunol. 5, 266-271. Bansal, A.S., et al., 2012. Mechanism of human chorionic gonadotrophin-mediated immunomodulation in pregnancy. Expert. Rev. Clin. Immunol. 8, 747-753. Bashiri, A., et al., 2018. Recurrent implantation failure-update overview on etiology,

ro of

diagnosis, treatment and future directions. Reprod. Biol. Endocrin. 16, 121. Borisova, M. A., et al., 2017. Human chorionic gonadotropin: unknown about known. Human Physiology, 43(1), 93-104., Moiseenko DY, Smirnova OV. Human

Chorionic Gonadotropin: Unknown about Known. Fiziol. Cheloveka. 43,97-110.

-p

Chang, A.L., et al., 2016. CCL2 Produced by the Glioma Microenvironment Is

Essential for the Recruitment of Regulatory T Cells and Myeloid-Derived

re

Suppressor Cells. Cancer Res. 76, 5671-5682.

Mol. Sci. 18, 2037.

lP

Chinedu, N., et al., 2017. hCG: Biological functions and clinical applications. Int. J.

Coughlan, C., et al., 2014. Recurrent implantation failure: definition and management.

na

Reprod. Biomed. Online. 28, 14-38.

Craciunas, L., et al., 2018. Intrauterine administration of human chorionic gonadotropin (hCG) for subfertile women undergoing assisted reproduction.

ur

Cochrane. Database. Syst. Rev. 10, Cd011537. Diao LH et al., 2017. Human chorionic gonadotropin potentially affects pregnancy

Jo

outcome in women with recurrent implantation failure by regulating the homing preference of regulatory T cells. Am. J. Reprod. Immunol. 77(3).

Du, M. R., et al., 2014. The integrative roles of chemokines at the maternal-fetal interface in early pregnancy. Cell. Mol. Immunol. 11, 438-448. Fatemeh, M., et al., 2017. Effect of intrauterine injection of human chorionic gonadotropin before embryo transfer on pregnancy rate: a prospective randomized study. J. Res. Med. Sci. 22, 6.

Gao M et al., 2019. Intrauterine injection of human chorionic gonadotropin before embryo transfer can improve in vitro fertilization-embryo transfer outcomes: a meta-analysis of randomized controlled trials. Fertil. Steril. 112(1):89-97.e1. Ghaebi, M., et al., 2017. Immune regulatory network in successful pregnancy and reproductive failures. Biomed. Pharmacother. 88, 61-73. Ghaebi, M., et al., 2019. T cell Subsets in Peripheral Blood of Women with Recurrent Implantation Failure. J. Reprod. Immunol. 131, 21-29. Guerin, L. R., et al., 2011. Seminal fluid regulates accumulation of FOXP3+

ro of

regulatory T cells in the preimplantation mouse uterus through expanding the FOXP3+ cell pool and CCL19-mediated recruitment. Biol. Reprod. 85, 397-408.

He, Y. Y., et al., 2007. Regulation of C-C motif chemokine ligand 2 and its receptor in

gestation. Hum Reprod.;22, 2733-2742.

-p

human decidual stromal cells by pregnancy-associated hormones in early

Huang, P., et al., 2018. Effects of intrauterine perfusion of human chorionic

Reprod. Immunol. 79, e12809.

re

gonadotropin in women with different implantation failure numbers. Am. J.

lP

Jiang, R., et al., 2017. Abnormal ratio of CD57(+) cells to CD56(+) cells in women with recurrent implantation failure. Am. J. Reprod. Immunol. 78, e12708.

na

Lin, D. X., et al., 2005. Dependence of uterine cyclooxygenase2 expression on luteinizing hormone signaling. Biol. Reprod. 73, 256-60. Liu, H., et al., 2013. CAPN 7 promotes the migration and invasion of human

ur

endometrial stromal cell by regulating matrix metalloproteinase 2 activity. Reprod. Biol. Endocrinol. 11, 64.

Jo

Liu X et al., 2019. Intrauterine administration of human chorionic gonadotropin improves the live birth rates of patients with repeated implantation failure in frozen-thawed blastocyst transfer cycles by increasing the percentage of peripheral regulatory T cells. Arch. Gynecol. Obstet. 299(4):1165-1172. Osman A et al., 2016. The effect of intrauterine HCG injection on IVF outcome: a systematic review and meta-analysis. Reprod. Biomed. Online. 33(3):350-9. Paiva, P., et al., 2011. Human chorionic gonadotrophin regulates FGF2 and other

cytokines produced by human endometrial epithelial cells, providing a mechanism for enhancing endometrial receptivity. Hum. Reprod. 26, 1153-1162 Panek, W.K., et al., 2019. Local Application of Autologous Platelet-Rich Fibrin Patch (PRF-P) Suppresses Regulatory T Cell Recruitment in a Murine Glioma Model. Mol. Neurobiol. 56, 5032-5040. Pijnenborg, R., et al., 1983. Uteroplacental arterial changes related to interstitial trophoblast migration in early human pregnancy. Placenta. 4, 397-413. Ramhorst, R., et al., 2016. Decoding the chemokine network that links leukocytes

ro of

with decidual cells and the trophoblast during early implantation. Cell. Adh. Migr. 10, 197-207.

Ren, J., et al., 2019. Myeloid-derived suppressor cells depletion may cause pregnancy loss via upregulating the cytotoxicity of decidual natural killer cells. Am. J.

-p

Reprod. Immunol. 81, e13099.

Reshef, E., et al., 1990. The presence of gonadotropin receptors in nonpregnant

Endocrinol. Metab. 70, 421-430.

re

human uterus, human placenta, fetal membranes, and decidua. J. Clin.

lP

Schumacher, A., et al., 2009. Human chorionic gonadotropin attracts regulatory T cells into the fetal-maternal interface during early human pregnancy. J. Immunol.

na

182, 5488-5497.

Schumacher, A., et al., 2009. Human chorionic gonadotropin attracts regulatory T cells into the fetal-maternal interface during early human pregnancy. J. Immunol.

ur

182, 5488-5497.

Schumacher, A., et al., 2012. Blockage of heme oxygenase-1 abrogates the protective

Jo

effect of regulatory T cells on murine pregnancy and promotes the maturation of dendritic cells. PLoS One. 7, e42301.

Sha, J., et al., 2017. Alteration of Th17 and Foxp3(+) regulatory T cells in patients with unexplained recurrent spontaneous abortion before and after the therapy of hCG combined with immunoglobulin. Exp. Ther. Med. 14, 1114-1118. Srivastava, A., et al., 2013. Profiles of cytokines secreted by isolated human endometrial cells under the influence of chorionic gonadotropin during the

window of embryo implantation. Reprod. Biol. Endocrinol. 11, 116 Takenaka, M. C., et al., 2017. Tolerogenic dendritic cells. Semin. Immunopathol. 39, 113-120. Vasilev, G., et al., 2019. Secretory factors produced by adipose mesenchymal stem cells downregulate Th17 and increase Tregs in peripheral blood mononuclear cells from rheumatoid arthritis patients. Rheumatol. Int. 39, 819-826. Zhang, N., et al., 2009. Regulatory T cells sequentially migrate from inflamed tissues to draining lymph nodes to suppress the alloimmune response. Immunity. 30,

ro of

458-469. Zhang, T., et al., 2019. Intrauterine infusion of human chorionic gonadotropin before embryo transfer in IVF/ET cycle: The critical review. Am. J. Reprod. Immunol. 81, e13077.

-p

Zheng, M., et al., 2001. Expression and Localization of Luteinizing Hormone

Receptor in the Female Mouse Reproductive Tract1. Biol. Reprod. 64, 179–187.

Jo

ur

na

lP

re

Zhou, J., et al., 2012. An increase of treg cells in the peripheral blood is associated with a better in vitro fertilization treatment outcome. Am. J. Reprod. Immunol. 68, 100-106.

Figure legends

ro of

Fig 1. Endometrial Tregs was decreased in women with RIF. Photomicrographs of endometrial tissue samples from control women (a) and women with RIF (b) which were immunostained with Foxp3 are shown at 400×magnification. The number of Foxp3+ cells per section (c) for ten control women and nine women with RIF are

-p

shown in the scatter plots. Bar = 50 μm. RIF, recurrent implantation failure, **,

Jo

ur

na

lP

re

P<0.01

Fig 2. Intrauterine perfusion of hCG increased endometrial Tregs. Photomicrographs of endometrial tissue samples from women with RIF before intrauterine perfusion of hCG (a) and after intrauterine perfusion of hCG (b) which were immunostained with Foxp3 are shown at 400×magnification. The number of Foxp3+ cells per section (c) for women with RIF before intrauterine infusion of hCG (n=9) and after intrauterine infusion of hCG (n=9) are shown in the scatter plots. Photomicrographs of uteruses

from pseudopregnant mice with no surgery (d), pseudopregnant mice horn injected PBS (e) and pseudopregnant mice horn injected hCG (f) which were immunostained with Foxp3 are shown at 400×magnification. The number of Foxp3+ cells per section (g) for pseudopregnant mice with no surgery, pseudopregnant mice horn injected PBS and pseudopregnant mice horn injected hCG are shown in the scatter plots. Bar = 50

lP

re

-p

ro of

μm. *, P<0.05, **, P<0.01

na

Fig 3. LH/CG receptors were expressed in the cytoplasm and nucleus of endometrial stromal cells. Fluorescence images show the immunolocalization of LH/CG receptor

ur

in human endometrial stromal cells (hESCs) (a–c). Negative control shows

Jo

nonspecific fluorescence (d-f). Bar = 25μm.

ro of -p re lP na

Fig 4. Chemokine CCL2 increased with hCG stimulation in vivo and in vitro. The

ur

mRNA levels of CCL2, CCL3, CCL4, CCL5, CCL7 and CXCL10 with or without 100IU/mL hCG stimulation were measured by qRT-PCR (a). The concentration of

Jo

CCL2 in the supernatant of endometrial stromal cells with or without 100IU/mL hCG stimulation were detected by enzyme linked immunosorbent assay (ELISA) (b).The CCL2 protein expression in hESCs with hCG stimulation for 0, 4, 12, 24 and 48 hours (c and d) was detected by western blot. The CCL2 protein expression in hESCs with 0, 20, 40, 100 IU/mL hCG stimulation (e and f) was also detected by western blot. Photomicrographs of endometrial tissue samples from control women (g), women with RIF before intrauterine perfusion of hCG (h) and after intrauterine perfusion of

hCG (i) which were immunostained with CCL2 are shown at 400×magnification. The integrated optical density (IOD) to area ratios of CCL2 in control women (n=10), women with RIF before intrauterine infusion of hCG (n=9) and after intrauterine infusion of hCG (n=9) are shown in the scatter plots (j). Bar = 50 μm. *, P<0.05, **,

na

lP

re

-p

ro of

P<0.01

Fig 5. hCG promoted the recruitment of Tregs toward hESCs. The schematic diagram

ur

of a model for Tregs migration assary is shown in (a). The purity of the isolated CD4+CD25+ T cells was verified by flow cytometry (b). The number of Tregs

Jo

migrating to lower chamber was detected by flow cytometry in the control group (c), the siRNA group (d), the hCG group (e) and the siRNA together with hCG group (f), and the results are shown in the column graph (g). The number of Tregs migrating to lower chamber was detected by flow cytometry in the control group (h), the CCR2 antagonist group (i), the hCG group (j) and the CCR2 antagonist together with hCG group (k), and the results are shown in the column graph (l). *, P<0.05, **, P<0.01

Fig 6. The verification that CCL2 siRNA treatment resulted in a significant

ro of

downregulation of CCL2 in hESCs. The mRNA level of CCL2 measured by qPCR in hESCs showed that CCL2 significantly downregulated with 50nM CCL2 siRNA

stimulation (a). The protein levels of CCL2 in supernatant of hESCs with control

-p

siRNA stimulation and CCL2 siRNA stimulation were measured by ELISA and CCL2 was significantly downregulated in hESCs with CCL2 siRNA atimulation (b). *,

Jo

ur

na

lP

re

P<0.05, **, P<0.01

Table 1 Primers were used for q-PCR. Sequences (5´–3´)

Length (bp)

human CCL2 F

GCTCATAGCAGCCACCTTCATTC

23

human CCL2 R

GGACACTTGCTGCTGGTGATTC

22

human CCL3 F

ACAGTGTGTTTGTGATTGTTTGCTC

25

human CCL3 R

AGCCACTCGGTTGTCACCAG

20

human CCL4 F

CTGTGCTGATCCCAGTGAATC

21

human CCL4 R

TCAGTTCAGTTCCAGGTCATACA

23

human CCL5 F

CGCTGTCATCCTCATTGCTACT

22

human CCL5 R

TGTGGTGTCCGAGGAATATGG

human CCL7 F

CCTGGGAAGCTGTTATCTTCAA

22

human CCL7 R

TGGAGTTGGGGTTTTCATGTC

21

human CXCL10 F

TGAAATTATTCCTGCAAGCCAA

22

human CXCL10 R

CAGACATCTCTTCTCACCCTTCTTT

25

human 18S F

CGGCTACCACATCCAAGGAA

20

human 18S R

CTGGAATTACCGCGGCT

17

-p

re

lP na ur Jo

ro of

Genes

21

Table 2 Baseline characteristics of participants. RIF group (n=9)

P value

Age(year)

31(25-38)

33(26-37)

0.7

BMI(kg/m2)

23.0±4.5

23.5±3.7

0.8

Basal FSH(IU/L)

7.3±0.9

7.6±1.7

0.7

Basal LH(IU/L)

4.9±2.8

5.5±2.8

0.7

Antral follicle count

16.0±5.0

16.0±7.3

1.0

Times of embryo transfer

1.0±0.0

3.9±1.1

0.001

Number of transferred embryos

1.7±0.5

6.3±2.0

Jo

ur

na

lP

re

-p

ro of

Control group (n=10)

0.002