Infertile human endometrial organoid apical protein secretions are dysregulated and impair trophoblast progenitor cell adhesion

Abstract

Introduction: Embryo implantation failure leads to infertility. As an important approach to regulate implantation, endometrial epithelial cells produce and secrete factors apically into the uterine cavity in the receptive phase to prepare the initial blastocyst adhesion and implantation. Organoids were recently developed from human endometrial epithelium with similar apical-basal polarity compared to endometrial gland making it an ideal model to study endometrial epithelial secretions.

Methods: Endometrial organoids were established using endometrial biopsies from women with primary infertility and normal fertility. Fertile and infertile organoids were treated with hormones to model receptive phase of the endometrial epithelium and intra-organoid fluid (IOF) was collected to compare the apical protein secretion profile and function on trophoblast cell adhesion.

Results: Our data show that infertile organoids were dysregulated in their response to estrogen and progesterone treatment. Proteomic analysis of organoid apical secretions identified 150 dysregulated proteins between fertile and infertile groups (>1.5-fold change). Trophoblast progenitor spheroids (blastocyst surrogates) treated with infertile organoid apical secretions significantly compromised their adhesion to organoid epithelial cell monolayers compared to fertile group (P < 0.0001).

Discussion: This study revealed that endometrial organoid apical secretions alter trophoblast cell adhesiveness relative to fertility status of women. It paves the way to determine the molecular mechanisms by which endometrial epithelial apical released factors regulate blastocyst initial attachment and implantation.

Keywords: apical secretion; blastocyst attachment; endometrial organoids; infertile organoids; intra-organoid fluid.

Figure 1

Establishment of fertile and infertile human endometrial organoids in vitro. (A) Representative images of fertile and infertile organoids under culture. The scale bar represents 200 μm. (B) The immunolocalization of CDH6 (tight junction marker) and E-cadherin (epithelial marker) was examined in the matched endometrial glands and organoids from the same donor. Similar localizations were recorded for all markers examined. An IgG control was included in which the non-immune antibody of the same isotype (IgG) was substituted for each primary antibody at the same concertation. The scale bar represents 20 μm.

Figure 2

The expression of receptivity markers is dysregulated in infertile organoids after hormone treatments. (A) Schematic of hormone treatments for organoids. Passaged fertile and infertile organoids were cultured for 4 days in ExM to reform organoid structure before being subjected to E2 (to mimic proliferative phase) or E2+MPA+cAMP treatment (to mimic receptive phase). (B, C) After hormone treatments, organoid cells were collected for RT-qPCR analysis to determine the expression of receptivity markers. Expression levels were normalized to 18S (n=7-9 biological replicates). All data were presented as mean ± SEM. *P<0.05, **P<0.01, ns, no significant difference.

Figure 3

Fertile endometrial organoid sourced epithelial cells and HEECs respond similarly to miR-29c overexpression. (A) Endometrial epithelial cells outgrew similarly from endometrial glands and organoids. The scale bar represents 100 μm. (B, C) Immunocytochemistry staining of organoid derived epithelial monolayer with E-cadherin (epithelial marker) and COL4A1 (basal membrane marker). An IgG control was included in which the non-immune antibody of the same isotype (IgG) was substituted for each primary antibody at the same concertation. The scale bar represents 50 μm. (D) Overexpression of miR-29c in fertile organoid-derived epithelial monolayers significantly reduced HTR8/SVneo trophoblast spheroid adhesion compared to control. Similar effects on adhesion have been identified using HEECs¹¹. (E, F) After transfection, organoid sourced epithelial monolayers were also subjected to RT-qPCR analysis to determine the expression of potential miR-29c targets as previously examined in HEECs¹¹. Consistently, only COL4A1 expression was significantly decreased after miR-29c overexpression compared to control. miR-29c expression was normalized to U6. Gene expression levels were normalized to 18S. All data were presented as mean ± SEM (n=5-6 biological replicates). *P<0.05, **P<0.01, ns, no significant difference.

Figure 4

Incubation of IOF significantly changes the adhesive capacity of both HTPC and HTR8/SVneo spheroids. Fertile and infertile organoids were treated with E2+MPA+cAMP and IOF (A) which represent organoid apical secretions were collected to treat trophoblast cells before spheroid adhesion assay. (A) Representative images are presented to demonstrate that after being lightly vortexed, organoids were mildly disrupted to release IOF. The scale bar represents 100 μm. (B) Schematic of IOF treatment before spheroid adhesion assay. (C) HTR8/SVneo cells were treated with three different percentages of IOF 1%, 5% and 10% (v/v) and adhesion on Ishikawa cell monolayers tested. Only incubation with 10% (v/v) fertile IOF significantly improved HTR8/SVneo spheroid adhesion on Ishikawa monolayer compared to medium control. (D) After optimization, 10% (v/v) fertile or infertile IOF were used to treat HTPC spheroids and their adhesive capacity was determined on fertile organoid derived epithelial monolayer from receptive endometrium. Incubation of fertile or infertile IOF significantly changed the adhesive capacity of HTPC spheroids compared to medium control and 10% (v/v) fertile EOF control. A representative image of HTPC spheroids attaching to fertile organoid derived epithelial monolayer was shown, scale bar: 400 μm. All data were presented as mean ± SEM (n=3-4 biological replicates). *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001, ns, no significant difference. IOF, Intra-organoid fluid; EOF, Extra-organoid fluid.

Figure 5

Proteomic comparison of fertile and infertile IOF after E2+MPA+cAMP treatment. (A) Volcano plot depicting fold changes associated with differentially expressed proteins between fertile and infertile IOF. Thresholds of ±≥1.5-fold change (P < 0.05) were implemented. (B) Heat map of the significantly differentially expressed proteins between fertile and infertile IOF with at least 1.5-fold difference. Top 15 (C) KEGG pathways, (D) GO molecular functions and (E) cellular components (ranked by -Log[p-value]) assigned to proteins that are significantly lower in infertile IOF with at least 1.5-fold difference compared to fertile (also see Supplemental Tables 3 - 6 ). (n=5 biological replicates). IOF, Intra-organoid fluid.

Figure 6

Examination of candidate proteins identified by mass spectrometry. (A) Immunoblot validation of MUC5AC expression in fertile and infertile IOF. Equal amounts of protein were loaded for examination (n=6-11 biological replicates). (B, C) qPCR examination of the production of candidate proteins in organoid cells. Top 6 significantly decreased (B) and increased (C) proteins ranked by fold change in infertile IOF compared to fertile IOF were selected for analysis. Gene expression levels were normalized to 18S (n=5 biological replicates). All data were presented as mean ± SEM. *P<0.05, **P<0.01, ns, no significant difference; IOF, Intra-organoid fluid.