LGR5 as a therapeutic target for decidualization-based endometrial mesenchymal stem cells therapy in thin endometrium

Abstract

Background

Decidualization of endometrial mesenchymal stem cells (ESCs) is crucial for the implantation of the embryo. Thin endometrium (TE) can impair endometrial receptivity by affecting the decidualization of ESCs. However, how TE affects the decidualization of ESCs remains unclear.

Methods

Through multiple transcriptome analyses, we found that LGR5 plays a crucial role in the decidualization of ESCs. In vitro, siRNA was used to knock down LGR5 in ESCs, and decidualization was assessed via morphology, qRT-PCR, Western blot, and ELISA for decidualization marker (PRL, IGFBP-1). Colony forming assay, EdU, CFSE, and tube formation were used to evaluate proliferation and secretion. Signaling pathway components (β-catenin, AXIN2, PGR) were analyzed. In vivo, an adeno-associated virus (AAV) delivered shRNA to knock down Lgr5 in mouse uterine to assess endometrial receptivity and pregnancy rates. A TE mouse model was established and treated with decidualized endometrial mesenchymal stem cells (DSCs) or LGR5-knockdown DSCs to evaluate therapeutic efficacy.

Results

LGR5 was significantly upregulated during decidualization and downregulated in TE tissues. In vitro, LGR5 knockdown impaired DSCs’ morphological transformation, reduced decidualization, restored proliferation, but diminished secretory function and angiogenic potential. Mechanistically, LGR5 knockdown suppressed the β-catenin pathway and downregulated progesterone receptor (PGR). In vivo, Lgr5 knockdown reduced endometrial thickness, receptivity markers (Lif, Hoxa10, Pgr), and embryo implantation numbers. Crucially, transplanting LGR5-knockdown DSCs into TE mice resulted in inferior restoration of endometrial structure, reduced angiogenesis, and lower pregnancy rates compared to DSCs.

Conclusion

LGR5 plays an important role in ESCs decidualization and may contribute to the therapeutic efficacy of DSCs for TE repair, potentially through regulation of PGR via the β-catenin pathway. These results indicate that LGR5 represents a candidate target worthy of further investigation for developing function-restoring therapies in TE.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13287-026-05173-4.

Introduction

The endometrium is a highly regenerative tissue that undergoes growth and shedding, and is regulated by hormones [1]. In the natural ovulation cycle or embryo transfer cycle, an endometrial thickness of less than 7 mm is defined as TE [2, 3]. The pathogenesis of TE is complex and involves factors such as abortion, frequent intrauterine surgeries, and infections [4, 5]. TE impairs blood supply to the uterus and early fetus, resulting in lower pregnancy rates, higher early miscarriage rates, and increased risks of adverse perinatal outcomes [6]. Conventional treatments include hormone therapy, uterine perfusion with granulocyte colony-stimulating factor, and platelet-rich plasma (PRP) [7]. However, due to individual treatment responses and insufficient experimental evidence, these approaches often fail to yield optimal results [8, 9]. Therefore, improving endometrial thickness and restoring endometrial function in TE patients remains crucial.

Endometrial decidualization involves the transformation of ESCs into specialized secretory DSCs in response to hormonal changes, supporting embryo implantation and placental development [10, 11]. This process prepares the endometrium for a window of receptivity, essential for embryo implantation [12]. Disruptions in decidualization can impair endometrial receptivity, resulting in infertility, miscarriage, preterm birth, or even fetal death [13, 14]. Studies show that endometrial thickness is crucial for embryo implantation and pregnancy [15]. In TE, loss of the functional endometrial layer, reduced stem cell numbers, and decreased receptivity hinder embryo implantation and growth [16]. Thus, TE may cause decidualization disorders, affecting endometrial receptivity [17]. However, the molecular mechanisms underlying TE-induced decidualization disorders are still unclear.

Stem cell therapy has emerged as a promising treatment for various reproductive challenges. ESCs can differentiate into various cell types and proliferate widely within the body [18]. This makes them a recognized therapeutic strategy for endometrial repair and regeneration, as they can replace damaged cells and promote tissue regeneration [19]. Our experiments showed that ESCs derived from menstrual blood can be induced into DSCs, with superior therapeutic effects on TE compared to ESCs [20]. Therefore, decidualization may play a key role in stem cell-based repair of TE.

LGR5 (Leucine-rich repeat-containing G protein-coupled receptor 5) is a G-protein-coupled receptor and a target gene of the Wnt signaling pathway [21], it was first identified by Barker et al. as a marker for intestinal and tumor stem cells [22]. Studies suggest that the Wnt signaling pathway plays a crucial role in the proliferation, differentiation, repair, and embryo implantation of the endometrium [3, 23]. Whether LGR5 can restore TE fertility by regulating DSCs remains unclear.

In this study, we first identified LGR5 as differentially expressed between decidualized and non-decidualized endometrial stem cells, then constructed stem cell lines with LGR5 knockdown and verified the results both in vitro and in vivo models. The results showed that LGR5 knockdown inhibited the decidualization process of ESCs in vitro, reduced endometrial receptivity, and lowered pregnancy rate in vivo. Moreover, LGR5 knockdown in DSCs appeared to diminish their therapeutic effects on TE. These findings may offer insights that could inform the future development of therapeutic strategies for TE.

Methods and materials

Ethical approval and sample collection

The ESCs cell line used in this study was derived from previous research [24]. Endometrial tissues were obtained from volunteers aged 25 to 40 years who had no history of intrauterine infections, endometrial polyps, uterine fibroids, or other reproductive system diseases. None of the volunteers had received hormonal therapy within 3 months prior to sample collection. All samples were collected during the proliferative phase of the menstrual cycle via hysteroscopic procedures, and were immediately stored in saline on ice after collection. The sample collection was approved by the Ethics Committee of the First Affiliated Hospital of the University of Science and Technology of China (USTC) (2021KY015) and conducted in accordance with the Declaration of Helsinki. All participants gave written informed consent.

The study used Balb/c female mice, aged 6 weeks, with body weights ranging from 20 to 30 g. The mice were purchased from Shanghai SLAC Laboratory Animal Company. The mice had free access to water and food and were housed in a controlled environment with a 12-hour light/dark cycle, at 25 ± 2 °C and 75–80% humidity. All animal procedures were approved by the Animal Care and Use Committee at the First Affiliated Hospital of USTC (2022-N(A)-119). The study has been reported in line with the ARRIVE guidelines 2.0 [25].

ESCs culture and decidualization differentiation

Primary ESCs were obtained from three independent healthy donors (aged 28, 30, and 31 years). The cells used in this study were derived from a well-characterized cell line established in our previous publication [24]. ESCs were cultured in Petri dishes with DMEM/F12 (Thermo Fisher, USA) supplemented with 10% FBS (Sigma, USA) and 1×Penicillin-Streptomycin Solution (Biosharp, China). ESCs within passages 3 through 6 were utilized in this study [26]. When growth reached 80%, a complete culture medium containing 17-β-E2 (10nM, Sigma, USA), MPA (1µM, Absin, China), and cAMP (0.5mM, Sigma, USA) was added to induce decidualization [27, 28]. After 3 days, the cell supernatants were collected, and prolactin (PRL) levels were measured using enzyme-linked immunosorbent assay (ELISA). The expression of decidualization-related markers, insulin-like growth factor-binding protein 1 (IGFBP-1) and PRL, was assessed using western blotting (WB) and reverse transcription quantitative PCR (RT-qPCR) [29]. All key in vitro experiments were performed using cells from three independent ESC lines.

Identification of differential genes

Total RNA was extracted from ESCs and DSCs, respectively, and RNA-sequencing (RNA-seq) and results analysis were conducted by Gene Denovo Biotechnology Co., Ltd (Guangzhou, China). An average of 10 million raw reads were generated each sample. After quality control using Fastp, the clean reads obtained were aligned to the human reference genome using HISAT2. Gene expression levels were then normalized to FPKM (Fragments Per Kilobase per Million mapped reads) values. Bioinformatics analysis was performed using R software (version 4.2.2). Principal component analysis (PCA) was conducted using the built-in R function. The raw p-values obtained from differential expression analysis were adjusted for multiple testing using the Benjamini-Hochberg method to control the false discovery rate (FDR). Genes with an FDR-adjusted p-value < 0.01 and absolute fold change > 1 were considered statistically significant differentially expressed genes (DEGs). To refine the analysis and link the results of sequencing of endometrial samples, we combined RNA-seq datasets from normal endometrium (NE) and TE to conduct a comprehensive analysis of DEGs from both groups, and Venn diagrams were generated for visualization [30]. Gene Set Enrichment Analysis (GSEA) was used to analyze the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways associated with the DEGs. A protein-protein interaction (PPI) information was predicted using the Search Tool for the Retrieval of Interacting Genes (STRING) online database (http://string-db.org), and DEGs in the enriched pathways were analyzed in detail.

Transfection of siRNA and administration of β‑catenin agonist

When ESCs reached 70% confluence, the complete culture medium was replaced with DMEM/F12 and then cultured for 24 h. Lipofectamine 3000 (Thermo, USA) and siRNA (20 µM, Sangon Biotech, China, the sequence is shown in Table S1) were each diluted separately in Opti-MEM (Thermo Fisher, USA), combined, and incubated at room temperature for 20 min to allow transfection complex formation. After ESCs transfection, the complete medium was replaced with decidual differentiation medium for further culture. To assess transfection efficiency, we used GFP-labeled ESCs and Cy-3-labeled siRNAs (Sangon Biotech, China), and counted intracellular fluorescent siRNAs. Each group conducts three replicates of the experiment. To activate the β‑catenin pathway, CHIR99021 (Targetmol, USA) was added to the cell culture medium at a final concentration of 5 µM.

Colony forming assay (Colony)

ESCs, DSCs+siNC (D+siNC), and LGR5 knockdown DSCs (D+siLGR5) were plated in 6-well plates. After 3 days of culture, 4% paraformaldehyde (Beyotime, China) was added for 15-minute fixation. The plate was then washed with PBS (Servicebio, China) and stained with 0.1% crystal violet solution (Sigma, USA) for 15 min. After washing with PBS, images were taken and the differences between the different groups were examined. Each group conducts three replicates of the experiment.

Ethylenediurea (EdU)

The EdU Cell Proliferation Detection Kit (Beyotime, China) was utilized in this experiment. After the EdU solution was added to the cells for 3 h, 4% paraformaldehyde was added for cell fixation at room temperature. The cells were then treated with 0.5% Triton X-100 (Servicebio, China) and incubated at room temperature for 12 min. Nucleus was stained with Hoechst 33,342 (Biosharp, China), the images were captured using a fluorescence microscope in a darkroom, and the percentage of positive cells. Each group conducts three replicates of the experiment.

Carboxyfluorescein succinimidyl ester (CFSE)

The CFSE Cell Division Tracker Kit (Elabscience, China) was used for this experiment. ESCs, D+siNC, and D+siLGR5 were incubated following the protocol provided according to the instructions. After incubation, the cells were cultured for 3 days. Following resuspension in PBS, cell proliferation was measured by flow cytometry (BD, FACSFortessa), and the MFI (Mean Fluorescence Intensity) were analyzed with FlowJo software. Each group conducts three replicates of the experiment.

Angiogenesis promotion assay

Pre-cooled matrix gel (Corning, USA) was evenly spread in a 48-well plate and incubated until fully solidified. Human umbilical vein endothelial cells (HUVECs, Oricell, China) were seeded on the Matrigel, and cell supernatants collected from different groups (ESCs, D+siNC, and D+siLGR5) were added. After 12 h, total branch points and length were observed under a microscope, and the results were analyzed using ImageJ software. Each group conducts three replicates of the experiment.

ESCs proliferation and migration

When 2 × 10⁶ ESCs were cultured 2 days and reached 90% confluence, marking lines were drawn on the back of a 6-well plate. A 200 µL pipette tip was used to create scratches perpendicular to the marking lines, and then washed with PBS. Supernatants collected from different groups (ESCs, D+siNC, and D+siLGR5) were added, and cultured for 24 h. Images were captured at 0 and 24 h, and the cell migration rate was analyzed using ImageJ software. Each group conducts three replicates of the experiment.

Enzyme-linked immunosorbent assay (ELISA)

Supernatants from different groups (ESCs, D+siNC, and D+siLGR5) were collected, and prolactin (PRL) levels were quantitatively analyzed using an ELISA kit (Proteintech, China), following the manufacturer’s instructions. Absorbance was measured using a microplate reader (Thermo, SpectraMax iD3), and PRL concentrations were determined by comparing sample values with a standard curve. Each group conducts three replicates of the experiment.

Protein extraction and western blot

Proteins were extracted as previously described [31]. The extracted proteins were loaded onto a pre-prepared FuturePAGE™ 4–20% gel (ACE Biotechnology, China) for electrophoresis and transferred onto a nitrocellulose (NC) membrane (Biosharp, China). After blocking with 5% skim milk powder (Biosharp, China) for 1 h, the primary antibody was added and incubated overnight at 4 °C. Afterward, the appropriate secondary antibody was added and incubated for 90 min. Following incubation, the NC membrane was exposed to the developer (Beyotime, China) and analyzed with a chemiluminescence imager. Gray values were quantified using ImageJ software (Primary and secondary antibody details are in Table S2). Each group conducts three replicates of the experiment.

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was extracted from cells or tissue using the RNA-quick Purification Kit (Esscience, China). cDNA was synthesized using the RT purification kit (Vazyme, China). Amplification was then carried out using the Power Track SYBR Green Master Mix (Thermo, USA). Target genes were determined using the LightCycler 96 Instrument. (Roche, Primer sequences used in the experiment are shown in Table S3). Each sample was amplified in duplicate for 40 cycles. Relative gene expression levels were determined using the ΔΔCt method and normalized to the housekeeping gene GAPDH. Each group conducts three replicates of the experiment.

Immunofluorescence (IF)

Paraffin-embedded uterine tissue Sect.  (5 μm thick) were deparaffinized in xylene solution (Sangon Biotech, China) and rehydrated through a graded ethanol series (100%, 90%, 80%, 70%, 60% ethanol; 10 min each). Antigen retrieval was performed using citrate antigen retrieval solution (pH = 6.0, Servicebio, China) for 40 min, sections were blocked with 5% normal goat serum in PBS for 1 h at room temperature. Primary antibodies were incubated overnight at 4 °C, followed by secondary antibody incubation for 1 h at room temperature. Nuclei were stained with DAPI (10 µg/ml, Absin, China) for 5 min and all incubation steps were followed by three 5-minute washes with PBS. Images were captured using a Zesis, Axio Observer7 fluorescence microscope and analyzed by ImageJ. For the statistical analysis of the immunofluorescence staining results, the n value represents the biological replicate. Each biological replicate is obtained by averaging the results of three technical replicates.

Establishment of Lgr5 knockdown mouse model

In this study, knockdown of mouse Lgr5 was achieved using an adeno-associated virus (AAV) vector. The recombinant AAV (serotype 9) was packaged with an expression cassette encoding a short hairpin RNA (shRNA) targeting mouse Lgr5 (sequence as in Table S1) driven by a U6 promoter. Viral particles were produced in HEK293T cells via triple plasmid transfection, purified by iodixanol gradient centrifugation, and titrated by qPCR, yielding a final titer of 1 × 10¹² vg/mL.

Mice were anesthetized via intraperitoneal injection of Zoletil 50 (10 mg/kg). The experiment randomly consisted of four groups: control endometrium (C), control endometrium after AAV injection (C + AAV), pregnant endometrium after AAV injection (P + AAV), and pregnant endometrium after AAV-Lgr5 injection (P + AAV-siLgr5). Fifteen mice per group, a total of 60 mice, were included in the experiment. Mice in all groups were anesthetized via intraperitoneal injection of Zoletil 50 (10 mg/kg). After anesthesia, the abdomen was disinfected with iodophor, and both uteruses were then exposed. In Group C, 25 µL of normal saline was injected into the uterus through a syringe. In the Group C + AAV and P + AAV, the same volume of AAV was injected, while in the Group P + AAV-siLgr5, the same volume of AAV-siLgr5 was injected. After the injection, the abdominal muscle and skin were sutured with 5 − 0 absorbable suture.

Two weeks after surgery, female mice were housed with healthy male mice at a 2:1 ratio, and vaginal plug formation was checked the following morning. The presence of vaginal plugs was considered Day 0.5 of gestation, mice were euthanized by cervical dislocation after deep anesthesia, and the uterine tissues were detected by WB and qPCR, 12 mice were sacrificed on the 14th day of gestation to count the number of embryos.

TE model and cell therapy TE models

Mice were anesthetized via intraperitoneal injection of Zoletil 50 (10 mg/kg, France). The abdomen was disinfected with iodophor, and both uteruses were then exposed. Inject 100 µL of 95% ethanol above the uterine bifurcation along the long axis of the uterus. The ethanol remained in the uterus for 1 min, then was rinsed twice with normal saline [32]. The contralateral uterus was treated in the same manner.

The cell therapy TE mouse model was based on the TE model and randomly divided into four groups: control (Group C), TE model (Group TE), TE model with DSCs treatment (Group TE+DSCs), and TE model with siLGR5-DSCs treatment (Group TE + D-siLGR5). Fifteen mice per group, a total of 60 mice, were included in the experiment. Group C received no treatment for ethanol-induced uterine injury, with only 50 µL of normal saline injected into the uterine cavity. In the TE+DSCs and TE + D-siLGR5 groups, after ethanol injury, 1 × 10⁶ cells were injected into the uterine cavity and then sutured with an absorbable suture. Two weeks post-operation, 3 mice were sacrificed for uterine tissue collection, and 12 mice were housed with male mice as described previously, and mice were euthanized by cervical dislocation after deep anesthesia between Days 12–16 of gestation to count the number of embryos.

Statistical analysis

Data were analyzed using GraphPad Prism 9.0. All results were presented as mean ± SEM. Student’s t-test was used for comparisons between two groups, while one-way ANOVA was applied for multiple group comparisons The n value denotes the number of biological replicates. The data point for each biological replicate is derived from the average of three independent technical measurements. A P-value of less than 0.05 was considered statistically significant.

Result

ESCs Decidualization induction and verification

As previously reported, the ESCs used in this experiment have been well validated for their mesenchymal stem cell characteristics [24]. Immunostaining confirmed positivity for SUSD2 and vimentin (VIM), further supporting their identity [33] (Figures S1). To induce differentiation of ESCs into DSCs, the culture medium was supplemented with E2, MPA, and cAMP for 3 days (Figs. 1A). Following induction, we observed obvious morphological changes. While ESCs maintained a typical elongated, spindle-like shape, DSCs appeared larger and more rounded, which is characteristic of decidualization (Figs. 1B). qRT-PCR analysis revealed that the mRNA levels of the decidual markers PRL and IGFBP-1 were significantly higher in the DSCs group compared to the ESCs group (Fig. 1C). Additionally, Western blot analysis confirmed that the protein expression of IGFBP-1 increased in the DSCs following decidualization (Fig. 1D). The concentration of PRL in the supernatant was also higher in the DSCs group than in the ESCs group (Fig. 1E). Immunofluorescence analysis revealed higher fluorescence intensity for IGFBP-1 in the DSCs compared to the ESCs (Fig. 1F).

Fig. 1.

ESCs were induced into DSCs in vitro and functional verification. A Flow chart illustrating the induced decidualization process, CM: ESCs culture medium. B The morphological differences between ESCs and DSCs, scale bar = 200 μm. C qRT-PCR analysis of PRL and IGFBP-1 mRNA expression in ESCs and DSCs. D Western Blot tested the expression levels of IGFBP-1 in ESCs and DSCs. E ELISA assay measuring PRL concentrations in the supernatants of ESCs and DSCs. F IF assay showing IGFBP-1 protein expression in ESCs and DSCs, blue for nucleus (DAPI), green for IGFBP-1, and red for mesenchymal cells (VIM), scale bar = 100 μm. G Colony Forming Assay showing the colony formation of ESCs and DSCs. H The CFSE experiment detects the fluorescence intensity (FITC) in ESCs and DSCs. All measurement data were expressed as mean ± SEM, with n = 3, **** P < 0.0001 were considered statistically significant, Student’s t test

Previous studies indicated that after decidualization, ESCs transition from a proliferative state to a secretory state, supporting embryonic implantation, and are associated with a reduction in their proliferative capacity [34, 35]. To assess the functional changes of the induced DSCs, we performed cell colony and CFSE-based proliferation tracer assays. The colony experiment showed that the ESCs group formed more colonies compared to the DSCs group (Fig. 1G). In the CFSE assay, DSCs (MFI:1327) exhibited higher fluorescence intensity compared with ESCs (MFI:195), indicating fewer cell divisions and more uniform fluorescence distribution among descendant cells (Fig. 1H). In conclusion, we successfully induced the transformation of ESCs into DSCs in vitro and confirmed their decidualization characteristics, both morphologically and functionally.

LGR5 as the primary differentially expressed gene through multiple transcriptome analyses

Next, we performed transcriptome sequencing to identify the key DEGs associated with the decidualization process in ESCs. By sequencing and analyzing the genes of the ESCs group and the DSCs group, we screened out 1204 differentially expressed genes, including 647 up-regulated genes and 557 down-regulated genes (p-value < 0.01, fold change (FC) > 1 or FC < -1) (Fig. 2A). This analysis showed that DSCs predominantly express up-regulated genes following ESC decidualization. A secondary screening of the top 20 most significantly up-regulated genes identified key candidates such as LGR5, SFBP2, WNT4, and IGFBP-1 (Fig. 2B).

Fig. 2.

Multiple transcriptome analysis of differentially expressed genes. A Volcano plot of differentially expressed genes (DEGs) between DSCs and ESCs. B Heatmap analysis of the top 20 up-regulated DEGs (DSCs VS ESCs). C Venn diagram showing common DEGs between DSCs vs. ESCs and NE vs. TE. D Gene ontology (GO) enrichment analysis of the specific quantities of DEGs. E GO enrichment analysis of the proportion of DEGs. F Kyoto Encyclopedia of Genes and Genomes (KEGG) signaling pathway analysis, the proportion of DEGs. G KEGG signaling pathway analysis of the specific quantities of DEGs. H The protein-protein interaction (PPI) network was constructed by DEGs. I qRT-PCR analysis of LGR5 mRNA expression in NE and TE tissue

To refine our DEGs selection, we cross-analyzed these genes with previously published data from human endometrium tissue [30]. Comparison of DEGs between the ESCs/DSCs groups and those from TE and normal endometrium (NE) tissues identified 503 common DEGs (Fig. 2C). Gene ontology (GO) analysis revealed that these DEGs were mainly enriched in biological processes (BP) such as ossification, vascular system development, and cell adhesion (Fig. 2D, E). In terms of cellular components (CC), they were predominantly associated with the extracellular matrix. Molecular function (MF) analysis indicated that these genes are involved in signal transduction within the extracellular matrix. KEGG pathway analysis further highlighted enrichment in the Wnt signaling pathway and skeletal muscle system (Fig. 2F, G).

While previous studies have emphasized the role of the Wnt signaling pathway in TE [36], the relationship between the skeletal muscle system and the endometrium remains unclear. Therefore, we focused on DEGs within the Wnt signaling pathway. To narrow our selection, we constructed a protein-protein interaction (PPI) network using the STRING database (Fig. 2H). Integrating DEG analysis, KEGG pathway, and PPI data, we identified LGR5 as a key gene. Notably, LGR5 expression in TE tissues was significantly lower than in NE tissues (Fig. 2I).

LGR5 knockdown impairs the decidualization of ESCs in vitro

The role of LGR5 in decidualization remains unclear. To investigate its effect on decidualization, we constructed a small interfering RNA targeting LGR5 (siLGR5) and transfected it into ESCs. To confirm successful transfection, GFP-labeled ESCs were transfected with PBS, Cy3-labeled siNC, and Cy3-labeled siLGR5. Fluorescence microscopy images taken over 3 days showed successful siNC and siLGR5 internalization in ESCs, with localization in the cytoplasm. No fluorescence was detected in ESCs (Fig. 3A). Next, we measuring the percentage of Cy3-positive in ESCs, DSCs transfected with siNC (D+ siNC) and DSCs transfected with siLGR5 (D+ siLGR5) to quantified the transfection efficiency. The results showed that both siNC and siLGR5 groups exhibited significantly higher proportions of fluorescent cells compared to ESCs (Fig. 3B).

Fig. 3.

LGR5 knockdown inhibits ESCs decidualization and secretion. A Representative fluorescence images of the intracellular uptake of siRNA after 3 days transfected with PBS, siNC, and siLGR5, blue for nucleus (Hoechst), green for the cell, and red for CY3-labeled siRNA, scale bar = 100 μm. B Percentage of CY3 positive cells. C Bright-field morphology of ESCs, D+siNC and D+siLGR5, scale bar = 200 μm. D qRT-PCR analysis of LGR5, PRL, and IGFBP-1 mRNA expression after siNC and siLGR5 transfection. E Western Blot tested the expression levels of IGFBP-1 in ESCs, D+siNC, and D+siLGR5. F Quantitative analysis of protein level with β-actin as the control. G ELISA assay for the concentration of PRL in the supernatants of ESCs, D+siNC, and D+siLGR5. The measurement data were expressed as mean ± SEM, with n = 3. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 were considered statistically significant. Student’s t-test and one-way ANOVA

To investigate the role of LGR5 in ESCs decidualization, we induced decidualization for 3 days and performed cytomorphological analysis, qRT-PCR, Western blotting, and ELISA. Morphological observation revealed that D+siLGR5 group exhibited a smaller cell size and more elongated spindle-like shapes, compared to the D+siNC group (Fig. 3C), indicating obvious changes in cell morphology following LGR5 knockdown. qRT-PCR analysis showed that LGR5 expression in the D+siLGR5 group was significantly lower than in the D+siNC group. Additionally, expression levels of decidualization markers, PRL and IGFBP-1, were significantly reduced in the D+siLGR5 group compared to the D+siNC group, suggesting that LGR5 knockdown impaired the decidualization process (Fig. 3D). Western blotting confirmed that protein levels of IGFBP-1, a key decidualization marker, were elevated in the D+siNC group but reduced in the D+siLGR5 group (Fig. 3E and F). Similarly, ELISA analysis of PRL concentration in the supernatant showed reduced levels in the D+siLGR5 group (Fig. 3G). These results collectively indicate that LGR5 knockdown inhibits the decidualization process of ESCs in vitro.

LGR5 knockdown restores the proliferation function of DSCs, but reduces their secretion function

To further explore the effect of LGR5 on the proliferative function of DSCs, we performed colony formation, flow cytometry, and EdU incorporation assays. Both colony formation and EdU assays showed that the number of colonies and immunofluorescence intensity in the D+siNC group were significantly lower than in ESCs. In the D+siLGR5 group, colony formation and fluorescence intensity increased compared to the D+siNC group, though they remained lower than in ESCs (Figs. 4A–C). These results suggest that decidualization reduces ESCs proliferation, but LGR5 knockdown restores proliferative capacity. In the CFSE assay, fluorescence intensity in the D+siNC (MFI:985) was higher than in ESCs (MFI:472), suggesting reduced cell divisions with evenly distributed fluorescence among descendant cells. After LGR5 knockdown, the number of cell divisions increased, resulting in lower fluorescence intensity (MFI:696) compared to the D+siNC group (Fig. 4D). These findings support the results from the colony formation and EdU assays.

Fig. 4.

LGR5 knockdown affects the DSCs proliferation and secretion. A Colony forming assay showing changes in the proliferation of ESCs, D+siNC, and D+siLGR5, purple represents cell colonies. B IF staining for EdU in ESCs, D+siNC, and D+siLGR5, scale bar = 100 μm. C Statistical analysis of EdU-positive proportion (%) in ESCs, D+siNC, and D+siLGR5, n = 6. D Representative histograms of CFSE proliferation experiments in ESCs, D+siNC, and D+siLGR5. E The tubular structure images formed by the supernatants from ESCs, D+siNC, and D+siLGR5 after 12 h, scale bar = 200 μm. F Statistical analysis of the branch points. G Statistical analysis of the branch length. H The effects of supernatants from ESCs, D+siNC, and D+siLGR5 on the proliferation and migration of HUVEC, scale bar = 200 μm. I Quantitative analysis of the results of the HUVEC scratches experiment. The measurement data were expressed as mean ± SEM, with n = 3. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 were considered statistically significant. one-way ANOVA

Previous studies have shown that supernatants from DSCs can promote the proliferation and migration of Human Umbilical Vascular Endothelial Cells (HUVECs) [20, 37]. To evaluate the effect of LGR5 on DSC secretion function, we performed tube formation and scratch assays using supernatants from ESCs, D+siNC, and D+siLGR5 groups cultured with HUVECs. Supernatants from D+siNC enhanced tube formation, promoting the formation of more nodes, longer tubes, and more extensive tube structures. However, in the D+siLGR5 group, these parameters were significantly reduced (Fig. 4E–G), suggesting that LGR5 knockdown impairs tube formation capacity. Similar results were observed in the scratch assay, where the D+siNC group showed greater cell proliferation and migration toward the wound area. In contrast, supernatants from the D+siLGR5 group inhibited cell proliferation and migration (Fig. 4H, I). In conclusion, LGR5 knockdown enhances the proliferative capacity of DSCs but significantly inhibits their ability to form tubes, as well as cell proliferation and migration. These results suggest that LGR5 knockdown restores DSC proliferation but impairs their secretion function.

LGR5 regulates PGR expression through the β-catenin signaling pathway

Given the pivotal role of the progesterone receptor (PGR) in decidualization [38], we hypothesized that LGR5 modulates PGR via β-catenin signaling. To investigate whether LGR5 regulates PGR via the β-catenin pathway, we analyzed their expression after LGR5 knockdown. In D+siNC, β-catenin and its target AXIN2 were downregulated, while PGR was upregulated—a pattern characteristic of decidualization. Conversely, D+siLGR5 led to a partial recovery of β-catenin/AXIN2 but a significant decrease in PGR. Critically, treatment with the β-catenin agonist CHIR99021 not only increased β-catenin and AXIN2 but also restored PGR expression (Fig. 5A). Western blot analysis corroborated these findings (Fig. 5B, C). These data demonstrate that LGR5 regulates PGR expression through the β-catenin signaling pathway, and disruption of this pathway impairs decidualization.

Fig. 5.

LGR5 knockdown inhibits PGR expression through the β-catenin signaling pathway. A qRT-PCR analysis of AXIN2, β-catenin, and PGR mRNA expression in ESCs, D+siNC, D+siLGR5 and D+siLGR5 + CHIR99021. B Western Blot tested the expression levels of AXIN2, β-catenin, and PGR in ESCs, D+siNC, D+siLGR5 and D+siLGR5 + CHIR99021. C Quantitative analysis of protein level with β-actin as the control. The measurement data were expressed as mean ± SEM, with n = 3. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 were considered statistically significant. one-way ANOVA

Lgr5 knockdown impairs endometrial receptivity and reduces pregnancy rate in vivo

Previous studies have demonstrated that intracavitary injection of siRNA or adeno-associated virus (AAV) can effectively reduce gene expression [39, 40]. In this experiment, we used intracavitary AAV injection to investigate the effect of Lgr5 knockdown on endometrial receptivity (Fig. 6A). Fluorescence detection at 4- and 8-weeks post-injection confirmed stable transfer of adenovirus-encapsulated shRNA into the uterus. No fluorescence was observed in the control group, but both the AAV and AAV-siLgr5 groups exhibited fluorescence at both time points, indicating that AAV remained in the uterus for up to 8 weeks after injection (Fig. 6B).

Fig. 6.

Lgr5 knockdown in the mice endometrium and its effect on the endometrial receptivity. A Schematic diagram of the pattern after Lgr5 knockdown. B In vivo imaging of the mouse endometrium at weeks 4 and 8 after being transfected with AAV and AAV-siLgr5, n = 9. C Relative expression of Lgr5, Pgr, Lif, and Hoxa10 mRNA levels in the endometrium of mice in the control endometrium (C), control endometrium after AAV injection (C + AAV), pregnant endometrium after AAV injection (P + AAV), and pregnant endometrium after AAV-siLgr5 injection (P + AAV-siLgr5), n = 3. D Western blot analysis of PGR, HOXA10, and LIF protein levels in the endometrium of mice in the C group, C + AAV group, P + AAV group, and P + AAV-siLgr5 group, n = 3. E Quantitative analysis of PGR, HOXA10, and LIF protein levels with β-actin as the control, n = 3. F Number of embryos in the uterine horns of the C group, C + AAV group, P + AAV group, and P + AAV-siLgr5 group. G Statistics of the number of uterine embryos in the C group, C + AAV group, P + AAV group, and P + AAV-siLgr5 group, n = 12. The measurement data were expressed as mean ± SEM. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 were considered statistically significant. one-way ANOVA

Next, to assess the effect of Lgr5 on endometrial receptivity in vivo, we extracted RNA and WB from the control endometrium (C), control endometrium after AAV injection (C + AAV), pregnant endometrium after AAV injection (P + AAV), and pregnant endometrium after AAV-siLgr5 injection (P + AAV-siLgr5). Lgr5 expression in the P + AAV group was significantly higher than in the C group, C + AAV, and P + AAV-siLgr5 group, confirming the successful knockdown of the mouse gene by AAV-siLgr5 (Fig. 6C). Gene expression analysis revealed that, compared to the P + AAV group, the expression of endometrial receptivity markers Pgr, Lif, and Hoxa10 was reduced in the P + AAV-siLgr5 group (Fig. 6C). Western blot analysis corroborated these findings, showing consistent changes in PGR, LIF, and HOXA10 protein levels (Fig. 6D, E). These findings suggest that the Lgr5 knockdown in the mouse uterus inhibited the endometrial decidualization, ultimately reducing endometrial receptivity.

Meanwhile, in order to verify whether Lgr5 knockdown in the endometrium has an impact on the conception of mice in vivo, we compared the number of embryos among the different groups. According to the results, no significant difference in the number of embryos was observed between the pregnant (P) group and the P + AAV group (8.92 vs. 8.83), indicating that AAV itself does not affect the pregnancy. However, compared to the pregnant group, the number of embryos in the P + AAV-siLgr5 group was significantly reduced (8.92 vs. 5.0), with an increase in embryo absorption, indicating that Lgr5 knockdown lowered the pregnancy rate in mice (Fig. 6F and G). In conclusion, Lgr5 knockdown in vivo inhibited the endometrium decidualization, leading to a reduced pregnancy rate in mice.

Inhibiting the expression of LGR5 in DSCs reduces its therapeutic effect on TE

To assess the role of LGR5 in the treatment of TE mice with DSCs, we transplanted DSCs (TE + D) and LGR5 knockdown DSCs (TE + D-siLGR5) into TE mice to evaluate the therapeutic effect (Fig. 7A). HE and MASSON staining results showed that the TE group had disordered endometrial cell arrangement, reduced endometrial thickness (214.5 μm vs. 471.3 μm), fewer glands (8.3 vs. 28), and increased fibrosis (62.67% vs. 29%) compared to the control group (Fig. 7B–F). These findings confirm the successful establishment of the mouse TE model. After DSC treatment, the TE mice showed increased endometrial thickness (439.8 μm), a higher number of glands (28.67), and reduced fibrosis (31.83%), with endometrial thickness approaching the control group’s levels. In contrast, LGR5 knockdown DSCs showed no significant difference in restoring endometrial thickness (400.3 μm vs. 439.8 μm) compared to original DSCs, but they performed worse in restoring gland number (21.17 vs. 28.67) and reducing fibrosis (41% vs. 31.83%).

Fig. 7.

LGR5 knockdown reduces the therapeutic effect of DSCs on TE mice. A Schematic diagram of the cell therapy. B H&E staining of mouse uteri in control, TE, TE+DSCs(D), TE + D-siLGR5 group, scale bar = 200 μm, enlarge scale bar = 50 μm. C Statistical analysis of endometrial thickness in control, TE, TE+DSCs(D), and TE + D-siLGR5 group, n = 6. D Statistical analysis of the number of endometrial glands in control, TE, TE+DSCs(D), and TE + D-siLGR5 group, n = 6. E MASSON staining of mouse uteri in control, TE, TE+DSCs(D), and TE + D-siLGR5 group, scale bar = 200 μm, enlarge scale bar = 50 μm. F Statistical analysis of the endometrial fibrosis area in mice of different treatment groups, n = 6. G Images showing the number of embryos in the uteri of mice from different groups. H Statistics on the number of uterine embryos in different groups of mice, n = 12. The measurement data were expressed as mean ± SEM. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 were considered statistically significant. one-way ANOVA

To further investigate the effect on pregnancy outcomes, we conducted a pregnancy experiment. The number of embryos in the TE group was significantly lower than in the control group. Both types of stem cell transplantation improved the pregnancy rate of TE mice, with the TE+DSCs group showing a significantly higher embryo number than the TE + D-siLGR5 group (Fig. 7G and H). In conclusion, LGR5 knockdown in DSCs resulted in reduced recovery of endometrial glands, more fibrosis, and a lower pregnancy rate in TE mice compared to treatment with DSCs alone.

LGR5 knockdown in DSCs reduces endometrial angiogenesis during TE treatment

Cell proliferation and angiogenesis are critical processes in stem cell therapy for TE. To investigate these mechanisms, we used immunofluorescence to assess the expression of cell proliferation (PCNA) and angiogenesis (CD31) markers in uterine tissues after DSCs (TE+DSCs) and LGR5 knockdown DSCs (TE + D-siLGR5) treatment. After cell therapy, both the TE+DSCs and TE + D‑siLGR5 groups showed increased fluorescence intensities of PCNA and CD31 in uterine tissues compared to the TE group. Angiogenesis in the TE+DSCs group was comparable to that of the control group and higher than in the TE + D-siLGR5 group (Fig. 8A, C). These findings were further corroborated by immunofluorescence staining for VEGFA (Figure S2). However, there was no statistically significant difference in PCNA expression between the treatment groups and the control group (Fig. 8B, D). These results explain why no difference was observed in endometrial thickness between the TE+DSCs and TE + D-siLGR5 treatments, despite a notably higher number of embryos following DSC treatment. Enhanced angiogenesis, associated with increased VEGFA expression, likely contributes to improved endometrial receptivity and thereby promotes higher pregnancy rates.

Fig. 8.

LGR5 knockdown in DSCs reduces angiogenesis without affecting cell proliferation in TE treatment. A Representative images of IF staining of CD31, green for CD31, and blue for Hoechst, scale bar = 40 μm. B Representative images of IF staining of PCNA, green for PCNA, and blue for Hoechst, scale bar = 40 μm. C Statistical analysis of CD31-positive area. D Statistical analysis of PCNA-positive cells. The measurement data were expressed as mean ± SEM, with n = 6. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001 were considered statistically significant. one-way ANOVA

Discussion

This study provides valuable insights into the role of LGR5 in endometrial decidualization and its impact on TE repair. Transcriptomic analyses revealed lower expression of LGR5 in TE and ESCs compared to NE and DSCs. Subsequent experiments confirmed that the absence of LGR5 in TE impaired ESCs decidualization, markedly reducing DSC secretion function. Moreover, LGR5 knockdown appeared to alter PGR expression in the endometrium via the WNT/β-catenin signaling pathway, leading to dysfunction in TE decidualization. In vivo experiments showed that Lgr5 knockdown in mice resulted in impaired decidualization, reduced endometrial receptivity, and lower pregnancy rates. Importantly, when LGR5 knockdown in DSCs, the repair effect on TE was hindered. This confirms that LGR5 deficiency disrupts normal ESCs decidualization, weakening its repair capacity. These findings underscore the pivotal role of LGR5 in endometrial function and suggest its potential as a therapeutic target for TE.

Decidualization is a critical change in ESCs, involving both morphological and functional alterations in response to estrogen and progesterone fluctuations [41–43]. In this study, we induced ESCs to differentiate into DSCs in vitro, successfully establishing a decidualization cell model. TE is characterized by a relatively thin endometrium with inadequate blood supply, leading to insufficient nutrient and hormone delivery [16, 44]. This results in abnormal function and impaired embryo implantation, often accompanied by disrupted decidualization [45, 46]. We compared transcriptomic data from ESCs and DSCs with previous data from NE and TE. Cross-analysis identified LGR5 as a key regulator in this process. Notably, LGR5 was shown to influence the Wnt/β-catenin signaling pathway, which regulates cell proliferation, differentiation, and angiogenesis. The expression of LGR5 was directly linked to Wnt/β-catenin pathway activation, highlighting its role in TE repair.

In vitro, we used siRNA to knock down LGR5 expression [47]. To confirm successful transfection, we fluorescently labeled the siRNA with CY3, allowing direct visualization of siRNA uptake [48]. Further experiments demonstrated that LGR5 knockdown impaired ESC decidualization. Our results showed that the β-catenin pathway was inhibited during ESCs decidualization. This pathway plays a vital role in cell proliferation and tissue repair [49, 50]. After LGR5 knockdown, the pathway partially recovered, also inhibiting DSCs aging. Appropriate aging of DSCs is essential for successful decidualization, facilitating embryo implantation.

PGR, which mediates the endometrial response to progesterone, is crucial for ESCs decidualization [51, 52]. Our findings elucidate that LGR5 fine-tunes this process by calibrating β-catenin signal strength, sufficient signaling promotes -driven secretory differentiation, whereas attenuated signaling upon LGR5 knockdown fails to sustain adequate PGR levels, impairing secretion while permissively allowing proliferation. The direct link was confirmed by rescuing PGR expression in D-siLGR5 with the β-catenin agonist CHIR99021. Thus, LGR5 acts as a molecular rheostat, shifting β-catenin signaling output from a proliferative toward a secretory program; disruption of this switch underlies the diminished therapeutic efficacy observed with LGR5-deficient DSCs.

In vivo, we used AAV to knock down Lgr5 expression and GFP-labeled AAV to track retention time [53, 54]. After 8 weeks of transfection in the mouse uterus, Lgr5 knockdown was still detectable, confirming the stability of the Lgr5 knockdown model. Subsequent results showed a reduction in endometrial glands, and increased fibrosis in the Lgr5 knockdown group. These morphological changes indicate impaired tissue remodeling, leading to compromised endometrial receptivity. The uterus could not undergo the necessary structural changes for embryo attachment and implantation. Given that endometrial receptivity is crucial for successful pregnancy, these results underscore the vital role of Lgr5 in maintaining receptivity [55, 56]. Lgr5 inhibition also reduced the expression of Pgr and endometrial factors (Hoxa10, Lif), further confirming the damage to receptivity. Pregnancy experiments supported these findings, showing that Lgr5 knockdown hindered embryo implantation. Overall, these results highlight the essential role of Lgr5 in maintaining endometrial thickness and receptivity, which are key to reproductive health.

Currently, the treatment for TE primarily relies on hormonal therapy (such as estrogen), intrauterine infusion of granulocyte colony-stimulating factor, or PRP [57]. These treatments provide exogenous stimulation to promote endometrial thickness and restore receptivity [57]. While beneficial for some patients, their effectiveness remains limited, as the main causes of TE are the reduction in the number of endometrial basal layer stem cells and impaired endometrial receptivity [58]. Traditional therapies mainly focus on increasing endometrial thickness, but have limited effects on restoring the endometrial receptivity. In contrast, stem cell-based therapy shows significant promise in promoting both endometrial regeneration and receptivity [7, 59]. Therefore, stem cell therapy can be regarded as an emerging treatment method.

Our previous work has shown that DSCs derived from menstrual blood yield superior repair outcomes in the TE model than ESCs, highlighting the importance of DSC functional enhancement for therapeutic efficacy [20]. A key advance of this study is that we identified LGR5 as a key functional factor determining the therapeutic efficacy of DSCs. We used 95% ethanol to damage the endometrium and established a stable TE mouse model [60, 61]. After DSCs treatment, the thickness of the endometrium, glandular regeneration, and pregnancy rate were improved, which confirmed the regenerative potential of DSCs treatment. However, after transplanting DSCs with LGR5 knockdown, the therapeutic effect was reduced, and angiogenesis was decreased, which is crucial for endometrial regeneration and the restoration of receptivity [62]. Additionally, the pregnancy rate in the DSCs-siLGR5 group was lower than that of the control group, highlighting the decline in the ability of DSCs to improve endometrial receptivity when LGR5 is at low levels. This further demonstrates the important role of LGR5 in the receptivity of the endometrium and highlights its role in enhancing the therapeutic potential of stem cells in addressing reproductive issues such as TE.

In conclusion, our study underscores the crucial role of LGR5 in regulating decidualization and its impact on repairing TE. Our findings provide a foundation for understanding the pathogenesis of TE and the potential clinical application of stem cell treatments. However, a limitation of our work is the current lack of a highly specific antibody for immunohistochemical detection of endogenous LGR5, which precluded detailed in vivo localization and temporal profiling; future studies using emerging technologies will help resolve its spatial expression dynamics. Additional mechanistic dissection and rigorous preclinical safety evaluation are warranted before clinical translation. Nonetheless, our findings position LGR5 as a promising target for developing more effective, function-restoring therapies for refractory TE.

Abbreviations

TE

Thin endometrium

ESCs

Endometrial mesenchymal stem cells

DSCs

Decidualize endometrial mesenchymal stem cells

AAV

Adeno-associated virus

PGR

Progesterone receptor

LGR5

Leucine-rich repeat-containing G protein-coupled receptor 5

PRP

Platelet-rich plasma

PRL

Prolactin

IGFBP-1

Insulin Like Growth Factor Binding Protein 1

Author contributions

Conception and design: Kai Chen and Shengxia Zheng. Surgical endometrial tissue collection: Jingxin Wang, Collection and assembly of data, data analysis, and interpretation: Kai Chen, Xueqing Ning, Lu Zong, and Huiru Wang. Statistical analysis: Kai Chen, Huiru Wang, Xin Zhao, and Shengxia Zheng. Draft and revise the article: Kai Chen, Xueqing Ning and Shengxia Zheng. All authors contributed to the discussion and approved the final version of the manuscript for submission.

Funding

This work was supported by the National Natural Science Foundation of China (82201792, 81971339 and 82571878).

Data availability

The data used and analyzed during the current study are available from the corresponding author upon reasonable request. The sequencing reads for cell and human sample information that supports the findings of this study have been deposited, the accession codes are PRJNA673823 and PRJCA043023.

Declarations

Ethics approval and consent to participate

Human samples were obtained from the Department of Obstetrics and Gynecology, Center for Reproductive and Genetics, The First Affiliated Hospital of USTC, University of Science and Technology of China, Hefei, China. This study was approved by the Ethics Committee of First Affiliated Hospital of USTC (code no. 2021KY015, Title: Regulatory network of Wnt/β-catenin signaling pathway in menstrual blood stem cells repairing thin endometrial fertility. Date of approval: 27 Jan 2021), and all patients had signed informed consent forms before undergoing hysteroscopic surgery. All animal experiments were conducted in accordance with the Chinese Council on Animal Care guidelines. All the procedures were approved by the Institutional Animal Care and Use Committee at the First Affiliated Hospital of USTC (code no. 2022-N(A)-119, Title: Regulatory network of Wnt/β-catenin signaling pathway in menstrual blood stem cells repairing thin endometrial fertility. Date of approval: 9 May 2023).

Consent for publication

All authors gave consent for publication.

Competing interests

The authors declare no competing interests.