Expression and Function of Meflin in Human Endometrial Decidualization

This study was approved by the Ethics Committee of Gunma University.

This work was supported by a Grant‐in‐Aid for Scientific Research (22K19594) awarded to Akira Iwase by the Japan Society for the Promotion of Science, Japan.

To examine the cyclic changes in Meflin and αSMA expression in the normal endometrium, we performed immunohistochemical analysis of endometrial tissues obtained during the proliferative and secretory phases. Meflin expression in endometrial stromal cells was significantly higher during the secretory phase than during the proliferative phase. In contrast, the expression of αSMA, a myofibroblast marker, was higher during the proliferative phase than during the secretory phase (Figure  1A,B ). These findings suggest that Meflin and αSMA expression are inversely regulated during the menstrual cycle. Expression patterns of Meflin and αSMA in the human endometrium during the menstrual cycle. (A) Representative immunohistochemical staining of Meflin and αSMA in human endometrial stromal cells during the proliferative and secretory phases. Brown DAB staining indicates positive expression; nuclei are counterstained with hematoxylin (blue). Insets show higher magnification views of the stromal region. Scale bar: 300 μm; inset scale bar: 100 μm. (B) Quantification of immunostaining intensity using the Quick‐score system (combining intensity and proportion of positive cells; score range: 0–12) in proliferative phase ( n  = 8) and secretory phase ( n  = 9) endometrial samples. Box plots show the median (horizontal line), quartiles (boxes), and range (whiskers). Statistical significance was determined using the Wilcoxon test (* p  < 0.05, ** p  < 0.01). Treatment with 17β‐estradiol, progesterone, and dibutyryl cAMP (EPA) induced morphological changes characteristic of decidualization. While untreated control cells and cells treated with 17β‐estradiol alone maintained a spindle‐shaped morphology, EPA‐treated cells transformed into rounded, polygonal cells (Figure  2A ). To investigate the role of Meflin in this process, we established a stable iESC line that overexpressed Meflin. Successful overexpression was confirmed by western blotting using DsRed‐monomer‐transfected cells as controls (Figure  2B ). Meflin‐overexpressing cells adopted a polygonal morphology even in the absence of hormonal stimulation, resembling decidualized cells. Upon EPA treatment, control cells underwent the expected transformation from spindle‐shaped to polygonal. However, Meflin‐overexpressing cells exhibited minimal additional morphological changes in response to EPA, maintaining their polygonal shape (Figure  2C ). These results suggest that Meflin expression alone is sufficient to induce morphological features resembling those of decidualization. Morphological changes in endometrial stromal cells during decidualization and following Meflin overexpression. (A) Representative phase‐contrast images of iESCs treated with vehicle (control), 17β‐estradiol (E2, 10 ng/mL), or E2 plus progesterone (P4, 1 μg/mL) and EPA (0.5 mM) for 4 days. Scale bar: 100 μm. (B) Western blot confirmation of Meflin overexpression in stably transfected iESCs. Control cells were transfected with the DsRed‐Monomer vector (negative control, NC), while Meflin‐overexpressing (Meflin‐OE) cells were transfected with a Meflin expression vector. β‐actin served as the loading control. (C) Representative phase‐contrast images comparing the morphology of control (DsRed‐Monomer) and Meflin‐OE cells under untreated conditions and after EPA treatment for 4 days. Scale bar: 100 μm. To validate our in vitro immunohistochemical findings, we examined the effects of decidualization on Meflin and SMA expression in iESCs. Consistent with the in vivo observations, Meflin protein expression was upregulated in decidualized iESCs, while SMA expression was decreased compared to control and 17β‐estradiol‐treated cells (Figure  3A ). Next, we evaluated the SMA levels in Meflin‐overexpressing cells. Western blot analysis demonstrated that Meflin overexpression led to a marked reduction in SMA expression compared to that in control cells (Figure  3B ). To assess the effect of Meflin on decidualization at the molecular level, we measured IGFBP1 mRNA expression. In control cells, EPA treatment significantly upregulated IGFBP1 expression, confirming successful decidualization. In contrast, Meflin‐overexpressing cells exhibited reduced baseline IGFBP1 expression and an attenuated response to EPA (Figure  3C ). To further investigate the mechanism underlying IGFBP1 regulation, we examined the expression and localization of FOXO1, a transcription factor critical for IGFBP1 expression. In control cells, EPA treatment induced the nuclear accumulation of FOXO1 and decreased cytoplasmic FOXO1. However, these changes were attenuated in Meflin‐overexpressing cells (Figure  4 ). These findings indicate that Meflin overexpression appears to be involved in cell‐shape remodeling induced by decidualization stimuli; however, it is paradoxically associated with suppression of FOXO1 nuclear translocation, leading to reduced expression of classical decidualization markers. These findings suggest that Meflin may differentially regulate structural and transcriptional components of the decidualization process. Effects of decidualization and Meflin overexpression on protein and gene expression in endometrial stromal cells. (A) Western blotting of Meflin and αSMA expression in iESCs treated with vehicle (control), 17β‐estradiol (E2, 10 ng/mL), E2 plus progesterone (P4, 1 μg/mL), or E2 plus P4 and EPA for 4 days. β‐actin served as the loading control. (B) Western blot analysis comparing αSMA expression between control cells (DsRed‐Monomer) and Meflin‐OE cells under various treatment conditions. β‐actin served as a loading control. (C) Quantitative real‐time polymerase chain reaction analysis of IGFBP1 mRNA expression (a decidualization marker) in control and Meflin‐overexpressing cells under untreated, E2‐only, and EPA treatment conditions for 4 days. Values were normalized to 18S rRNA and expressed relative to control. Data are presented as the mean ± SD from three independent experiments. Statistical significance was determined using Student's t ‐test (* p  < 0.05, ** p  < 0.01, n.s.: Not significant). Subcellular distribution of FOXO1 and expression of Meflin and αSMA in control and Meflin‐overexpressing endometrial stromal cells under decidualization conditions. Western blotting of total cell lysate (TCL), nuclear (Nuc), and cytosolic (Cyto) fractions prepared from control cells (DsRed‐Monomer) and Meflin‐overexpressing cells (Meflin‐OE) treated with vehicle (control), 17β‐estradiol plus progesterone (E2 + P4), or E2 plus P4 and dibutyryl cAMP (EPA) for 4 days. TCL was probed for Meflin and aSMA with β‐actin serving as a loading control. Nuclear fractions were probed for FOXO1 with Lamin A/C serving as a nuclear fraction marker. Cytosolic fractions were probed for phosphorylated FOXO1 (pFOXO1) with β‐actin serving as cytosolic fraction marker. To determine whether the observed decrease in αSMA expression was functionally linked to reduced contractility, collagen gel contraction assays were performed. Decidualized iESCs exhibited significantly reduced contractile activity compared to control and 17β‐estradiol‐treated cells (Figure  5A,B ), consistent with decreased αSMA levels. Similarly, Meflin‐overexpressing cells showed significantly diminished gel contraction compared to control cells (Figure  5C,D ), mirroring their lower αSMA expression. These results support the functional role of Meflin in regulating contractility in endometrial stromal cells, likely through the modulation of αSMA expression. Assessment of cellular contractility using the collagen gel contraction assay. (A) Representative images of collagen gel contraction assay using iESCs treated with vehicle (control), 17β‐estradiol (E2, 10 ng/mL), or E2 plus progesterone and dibutyryl cAMP (EPA) for 2 days. Smaller gel area indicates greater contractility. (B) Quantification of gel contraction shown in (A). The graph shows the relative gel area (in pixels) compared to the initial size, analyzed using ImageJ software. Higher values indicate less contraction. (C) Representative images of collagen gel contraction assay comparing control cells (DsRed‐Monomer) and Meflin‐OE cells under different treatment conditions. (D) Quantification of gel contraction shown in (C). Data in (B) and (D) are presented as the mean ± SD from three independent experiments. Statistical significance was determined using Student's t ‐test (* p  < 0.05, ** p  < 0.01, n.s.: Not significant).

In this study, Meflin was identified as a novel and functionally important regulator of endometrial stromal cell differentiation during decidualization. Our findings demonstrated that Meflin modulates both the morphological and mechanical properties of stromal cells, particularly by suppressing αSMA expression and reducing cellular contractility. These effects were observed in both in vivo (menstrual cycle–dependent expression patterns) and in vitro models (hormone‐induced decidualization). This reduced contractile capacity is consistent with the phenotype of decidualized stromal cells, which exhibit decreased mechanical stiffness to support implantation and facilitate immune cell trafficking [ 16 , 17 ]. Interestingly, while Meflin overexpression promoted a morphological shift from spindle‐shaped to polygonal cells resembling a decidualized morphology, it paradoxically suppressed IGFBP1 and nuclear FOXO1 localization, a well‐established biochemical marker of decidualization. This dissociation underscores the complexity of decidualization, suggesting that morphological transformations and functional differentiation are governed by distinct molecular pathways. While IGFBP1 transcription is regulated by canonical progesterone/cAMP‐responsive transcription factors, such as FOXO1 and C/EBPβ [ 5 , 8 , 10 ], morphological remodeling is primarily mediated by cytoskeletal pathways involving Src–focal adhesion kinase–RhoA/Rho‐associated protein kinase signaling and myosin light chain phosphorylation [ 11 , 27 ]. Therefore, our data suggest that Meflin acts predominantly on the structural and mechanical axes of stromal cell differentiation, independently of hormone‐driven transcriptional programs. Moreover, our findings provide insights into FMT within the endometrium. αSMA is a canonical marker of myofibroblasts and FMT and is associated with increased contractility and tissue fibrosis [ 15 ]. In pathological conditions, such as endometriosis, stromal cells exhibit sustained myofibroblast‐like features, including elevated αSMA expression and enhanced contractility [ 15 , 28 ]. Our preliminary data suggest that Meflin is downregulated in endometriotic stromal cells (data not shown), supporting the hypothesis that Meflin serves as a negative regulator of pathological FMT [ 28 , 29 , 30 ]. The anti‐fibrotic role of progestins in endometriosis has been increasingly recognized in recent studies. Preoperative hormonal therapy, including progestins, has been shown to significantly reduce collagen type I deposition and TIMP‐1 expression in deep‐infiltrating endometriosis (DIE) lesions [ 31 ]. At the molecular level, progestin treatment upregulates decorin, a proteoglycan that inhibits TGF‐β activity and collagen fiber formation, and attenuates osteopontin expression in endometriotic lesions, both contributing to suppression of fibrogenesis [ 32 , 33 ]. A recent meta‐analysis further confirmed that long‐term progestin therapy not only slows the progression of fibrotic DIE lesions but also reduces lesion volume by approximately 28% [ 34 ]. These findings collectively support the disease‐modifying anti‐fibrotic effect of progestins beyond their traditional role in decidualization induction. Given that Meflin expression is upregulated during progesterone‐induced decidualization (as shown in this study) and that progestins exert anti‐fibrotic effects in endometriosis, we propose that Meflin functions as a downstream mediator of progestin‐driven suppression of pathological fibrosis. Clinically, these findings are relevant to reproductive disorders characterized by impaired decidualization and aberrant fibrosis. Although decidualization has traditionally been viewed as a progesterone‐dependent process [ 1 ], many cases of decidualization failure occur despite normal serum progesterone levels, implicating progesterone resistance as the central pathological mechanism [ 5 ]. In endometriosis, up to one‐third of patients exhibit progesterone resistance, which limits the efficacy of hormone‐based treatments [ 35 , 36 ]. Molecular contributors to this resistance include decreased expression of the progesterone receptor B isoform, epigenetic silencing of progesterone receptor genes, and alterations in downstream signaling pathways [ 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 ]. Our identification of Meflin as a progesterone‐responsive, antifibrotic factor provides a potential mechanistic link between hormonal resistance, stromal dysfunction, and reproductive failure. Taken together, restoration of Meflin, either through improved progesterone sensitivity or direct therapeutic targeting, may offer a novel strategy to reverse pathological fibrosis and restore the stromal remodeling capacity required for decidualization. Although our study elucidated the functional role of Meflin in decidualization, several limitations should be acknowledged. The upstream signaling pathways that regulate Meflin expression and the downstream mechanisms by which Meflin suppresses αSMA and influences cellular morphology remain unclear. Additionally, our immortalized stromal cells produced insufficient prolactin levels for reliable quantification, limiting our ability to fully characterize the functional decidualization response using multiple canonical markers. Further investigations are required to delineate how Meflin interacts with mechanical signaling cascades, whether its expression is regulated by hormonal or epigenetic mechanisms, and whether Meflin restoration can reverse progesterone resistance in endometriosis. In summary, Meflin was identified as a key regulator of endometrial stromal cell morphology and contractility during decidualization. Meflin overexpression promoted the morphological transformation characteristic of decidualization, while paradoxically suppressing functional decidualization markers, particularly nuclear FOXO1 localization and IGFBP1 expression (Figure  6 ). By modulating αSMA expression and suppressing contractile force, Meflin contributes to the structural transformation of the endometrial stroma into a receptive, non‐fibrotic environment. Its downregulation in endometriosis highlights its potential role in the pathogenesis of progesterone resistance and stromal dysfunction. Future studies should aim to identify the upstream regulators of Meflin and explore its therapeutic potential in treating reproductive disorders characterized by impaired decidualization and aberrant FMT. Proposed model of Meflin function in endometrial decidualization. Decidualization stimuli induce Meflin upregulation in endometrial stromal cells, which promotes morphological transformation and reduces αSMA but impairs functional decidualization by suppressing nuclear FOXO1 localization and subsequent IGFBP1 expression.

Decidualization is a progesterone‐dependent morphological and functional transformation of endometrial stromal fibroblasts that is essential for successful pregnancy [ 1 ]. This process can be recapitulated in vitro by treatment with estradiol (E2), progesterone (P4), and cyclic AMP (cAMP) [ 2 , 3 ]. During decidualization, endometrial stromal cells undergo marked changes, including cytoplasmic enlargement, glycogen accumulation, and enhanced secretory activity. Impaired decidualization has been implicated in various reproductive disorders, such as infertility, recurrent pregnancy loss, and placental dysfunction [ 4 , 5 ], and multiple rodent models have demonstrated a strong correlation between defective decidualization and infertility [ 6 ]. Recent studies have elucidated key transcriptional regulators of decidualization, including CCAAT/enhancer‐binding protein beta (C/EBPβ), forkhead box protein O1 (FOXO1), and Wilms tumor 1 [ 7 , 8 , 9 , 10 ]. These transcriptional networks coordinate the profound morphological and cytoskeletal changes characteristic of decidualizing stromal cells. In particular, decidualization is accompanied by dynamic cytoskeletal remodeling, such as actin filament reorganization, stress fiber disassembly, and nuclear actin remodeling [ 11 , 12 ]. Notably, plasminogen activator inhibitor‐1 has been shown to promote F‐actin reorganization during decidualization by suppressing a key signaling cascade involved in cell motility and proliferation [ 13 ]. Notably, alpha‐smooth muscle actin (αSMA), a hallmark of myofibroblasts, is significantly downregulated during decidualization [ 14 ]. Myofibroblasts are characterized by elevated αSMA expression, which directly contributes to their contractile function [ 15 ]. Tsuno et al. demonstrated that reduced αSMA expression during decidualization leads to diminished cellular contractility, which is essential for establishing a receptive and compliant uterine environment for implantation [ 14 , 16 , 17 ]. The transition between the fibroblast and myofibroblast phenotypes (fibroblast‐to‐myofibroblast transition, FMT) is regulated by diverse stimuli, including transforming growth factor beta 1 signaling, mechanical stress, substrate stiffness, and epigenetic mechanisms [ 18 ]. However, the molecular mechanisms governing this balance in the endometrium during decidualization remain poorly understood. Meflin, encoded by immunoglobulin superfamily containing leucine‐rich repeats (ISLR), is a glycosylphosphatidylinositol (GPI)‐anchored protein expressed in mesenchymal stromal/stem cells and tissue‐resident fibroblasts [ 19 ]. Meflin suppresses SMA expression and inhibits fibrosis in various physiological and pathological conditions. In pancreatic cancer stroma, Meflin expression is inversely correlated with αSMA levels, and its presence in cancer‐associated fibroblasts is associated with better patient outcomes and attenuated tumor progression [ 20 ]. Similarly, in colorectal cancer, Meflin modulates tissue architecture by enhancing bone morphogenetic protein signaling, contributing to the balance between stromal activation and suppression [ 21 ]. In non‐malignant tissues, Meflin‐positive fibroblasts have been shown to protect against excessive fibrosis in cardiac and pulmonary models [ 22 , 23 ]. Collectively, these findings suggest that Meflin is a key regulator of fibroblast phenotype and contractility. Given the inverse relationship between decidualization and FMT, and the established role of Meflin in repressing myofibroblast differentiation, we hypothesized that Meflin may regulate endometrial decidualization by modulating αSMA expression and cellular contractility. In this study, we investigated the expression patterns and functional roles of Meflin during decidualization to provide novel insights into the molecular mechanisms governing endometrial stromal remodeling. The results contribute to a better understanding of reproductive disorders characterized by impaired decidualization.

The authors declare no conflicts of interest.

Endometrial tissues were obtained from patients undergoing hysterectomy for early stage cervical cancer, grade 3 cervical intraepithelial neoplasia, borderline ovarian tumors, or uterine fibroids at Gunma University Hospital. Patients with endometriosis, adenomyosis, endometrial cancer, or other endometrial lesions were also excluded. The menstrual cycle phase was determined by histological dating using standard criteria. Formalin‐fixed, paraffin‐embedded tissue sections (2‐μm thick) were deparaffinized and rehydrated. Antigen retrieval was performed using Immunosaver (Nisshin EM, Tokyo, Japan) at 98°C for 45 min. Endogenous peroxidase activity was quenched using 0.3% hydrogen peroxide. After blocking with 10% bovine serum albumin, the sections were incubated overnight at 4°C with rabbit monoclonal anti‐ISLR antibody (1:400) (Atlas Antibodies, Bromma, Sweden) and rabbit monoclonal anti‐α‐smooth muscle actin antibody (1:1000) (#19245; Cell Signaling Technology, Danvers, MA, USA). Immunoreactivity was visualized using the EnVision+ system (K4003; Agilent, Santa Clara, CA, USA) and 3,3′‐diaminobenzidine. Sections were counterstained with Mayer's hematoxylin, mounted, and examined under a BX51 microscope (Olympus, Tokyo, Japan). Immunostaining was evaluated using the Quick‐score method [ 24 ], which incorporates both staining intensity and the proportion of positive cells. Immortalized endometrial stromal cells (iESCs) were established as previously described [ 25 ]. Cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% charcoal‐stripped fetal bovine serum (FBS), 50 U/mL penicillin, and 50 μg/mL streptomycin at 37°C in a humidified 5% CO 2 incubator. The cells were passaged at 80%–90% confluence using 0.25% trypsin–EDTA. The culture medium was changed every 3 days. To examine the effects of decidualization on Meflin expression, iESCs were seeded at 60% confluence in DMEM supplemented with 10% charcoal‐stripped FBS. After cell attachment, the cells were cultured under three conditions: (1) medium alone (control), (2) medium with 17β‐estradiol (E2; 10 ng/mL; Sigma‐Aldrich, St. Louis, MO, USA), and (3) medium with E2, progesterone (P4; 1 μg/mL; Sigma‐Aldrich), and N6,2′‐O‐dibutyryl cyclic AMP (cAMP; 0.5 mM; Sigma‐Aldrich), referred to as dibutyryl cAMP (EPA) [ 2 , 3 ]. The medium was replaced every 2 days, and the cells were harvested on day 4. Total RNA was extracted using the NucleoSpin RNA Plus Kit (Takara Bio, Kusatsu, Japan) and reverse‐transcribed using ReverTra Ace qPCR RT Master Mix (TOYOBO, Osaka, Japan). Quantitative real‐time PCR was performed using GeneAce SYBR qPCR Mix Low ROX (Nippon Gene, Tokyo, Japan) on an Eco Real‐Time PCR System (Illumina, San Diego, CA, USA). The primer sequences are listed in Table  S1 . The expression levels were normalized to those of 18S rRNA. Each sample was analyzed in triplicate and the mean value was used for statistical analysis. Cells were lysed in RIPA buffer containing a protease inhibitor cocktail (03969–21; Nacalai Tesque, Kyoto, Japan). Ten micrograms of total protein were separated on 10% SDS–PAGE gels and transferred to nitrocellulose membranes. Membranes were blocked with 5% bovine serum albumin and incubated overnight at 4°C with the following antibodies: rabbit monoclonal anti‐ISLR (1:2000) (HPA050811; Atlas Antibodies), mouse monoclonal anti‐α‐smooth muscle actin (1:5000) (A5228; Sigma‐Aldrich), rabbit monoclonal anti‐FoxO1 (1:1000) (#2880; Cell Signaling Technology), rabbit polyclonal anti‐phospho‐FoxO1 (Ser256) (1:1000) (#9461; Cell Signaling Technology), mouse monoclonal anti‐Lamin A/C (1:5000) (#2032; Cell Signaling Technology), and mouse monoclonal anti‐β‐actin (1:5000) (A5316; Sigma‐Aldrich). After incubation with HRP‐conjugated goat anti‐mouse IgG (1:1000–5000) (SA00001‐1; Proteintech, Rosemont, IL, USA) or goat anti‐rabbit IgG (1:1000–5000) (SA00001‐2; Proteintech), bands were visualized using Clarity Western ECL substrate (1705060; Bio‐Rad, Hercules, CA, USA) and detected with the ImageQuant LAS500 system (GE Healthcare, Tokyo, Japan). Nuclear and cytoplasmic proteins were isolated using NE‐PER Nuclear and Cytoplasmic Extraction Reagents (78833; Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's protocol. A plasmid encoding the ISLR was constructed via multi‐step cloning. The ISLR cDNA subcloned into the pEX‐A2J2 vector was obtained from Eurofins Genomics (Tokyo, Japan). Both the pEX‐A2J2‐ISLR and pCMV‐MycC vectors were digested with EcoRI‐HF and SalI‐HF (New England Biolabs, Ipswich, IL, USA) and ligated using T4 DNA ligase to generate pCMV‐ISLR‐MycC. Assembly was confirmed by DNA sequencing (Eurofins Genomics) using the primers listed in Table  S1 . Alternatively, fragments were assembled using the NEBuilder HiFi DNA Assembly (E2621X; New England Biolabs) and validated by additional sequencing (Table  S2 ). An empty pDsRed‐Monomer vector was used as a control. i ESCs were transfected with plasmids using ViaFect (Promega, Madison, WI, USA) and selected with 100 μg/mL hygromycin. Overexpression was confirmed by PCR and western blotting. Collagen gel contraction assays were performed as previously described [ 26 ]. i ESCs (5.0 × 10 5 cells) were embedded in type I‐A collagen (Nitta Gelatin, Osaka, Japan) and placed in 24‐well plates. After polymerization, the gels were overlaid with medium containing 10% charcoal‐stripped FBS with or without hormonal treatment. The gels were floated and incubated for 2 days, and the gel surface area was quantified using ImageJ software (NIH, USA). All experiments were independently repeated at least three times. Data are expressed as the mean ± standard deviation (SD). Statistical analyses were performed using R version 4.2.2. For immunohistochemical analysis, group differences were assessed using the Wilcoxon test. For quantitative PCR and gel contraction data ( n  = 3 per group), unpaired Student's t ‐test was used. Statistical significance was set at p  < 0.05.

Table S1: Primers for vector construction and sequencing. Table S2: Primers for vector construction and sequencing.