miRNA expression data in UCEC were downloaded from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/). A cohort comprising 578 samples were analyzed, with 33 categorized as normal and 545 as UCEC tumor samples. Additionally, 49 DE-miRNAs previously identified by Zhou et al (21) from the comparison of 56 plasma samples from patients with endometrial cancer and healthy controls were analyzed in the present study.

The Limma package (version 3.52.0; Bioconductor) was employed for the analysis of DE-miRNAs between UCEC and normal samples. miRNAs with a false discovery rate ≤0.05, |log₂fold change|≥1.2 and P-value <0.01 were considered to be statistically significant. The 49 DE-miRNAs identified from analysis by Zhou et al (21) were included in the present study analysis.

From October 2023 to April 2024, a total of 9 patients (age, 44–72 years; mean age, 57.56±8.93 years) diagnosed with UCEC via pathological examination were enrolled in Renmin Hospital of Wuhan University (Wuhan, China). Inclusion criteria: Patients had been admitted to the hospital for the first time and did not receive any malignant tumor treatments, such as chemotherapy or radiotherapy before surgery; postoperative tissues were diagnosed by pathologists as UCEC with parallel International Federation of Gynecology and Obstetrics (FIGO) staging; patients were aware of and agreed to the present study process; and patient clinical data and follow-up data were complete and available. Exclusion criteria: Patients with other malignant tumors; patients with immune system diseases or infectious diseases; and patients with severe impairment of the heart, liver, kidney and other organ functions. According to the FIGO staging system (22), the tumor stages of the included patients were as follows: Stage I, 1; stage III, 1; and stage IV, 7 cases. A total of 9 sex and age-matched healthy individuals were recruited as normal controls (age, 41–70 years; mean age, 55.42±7.28 years). Peripheral blood samples (10 ml) from both patients with UCEC and healthy individuals were collected, followed by centrifugation at 3,000 × g for 10 min at 4°C within 4 h to obtain plasma supernatant. The collected plasma samples were further subjected to centrifugation at 16,000 × g for 10 min at 4°C and subsequently stored at −80°C until further use. The present study adhered to the principles outlined in the Declaration of Helsinki, and ethical approval was obtained from Renmin Hospital of Wuhan University Ethics Committee (approval no. WDRY2023-K161). All participants signed a written informed consent form.

Procell Life Science & Technology, Ltd. provided two UCEC cell lines (HEC-1-A and Ishikawa). HEC-1-A and Ishikawa cells were grown in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.), with an environment with 5% CO₂ and a temperature of 37°C. Cells in the logarithmic growth phase were used for the subsequent experiments.

miR-32-5p-inhibitor, miR-32-5p inhibitor-negative control (NC), overexpression (OE)-Forkhead Box N2 (FOXN2), OE-NC, miR-32-5p-mimic and miR-32-5p-mimic-NC were synthesized by Ribo Biotech, Ltd. The vector used for the overexpression transfections was pcDNA^™3.1⁽⁺⁾ (Invitrogen; Thermo Fisher Scientific, Inc.). Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.), HEC-1-A cells were transfected with miR-32-5p-inhibitor or miR-32-5p inhibitor-NC, while Ishikawa cells were transfected with miR-32-5p-mimic or miR-32-5p-mimic-NC. The concentration of nucleic acid used was 50 nM; the transfection process lasted for 48 h at 37°C, after which the cells were collected further functional experiments. The sequences of the miRNA mimic and inhibitors as well as those of the negative controls are listed in Table I.

Exosomes from UCEC plasma samples (UCEC-Exo) and normal controls (Normal-Exo) were isolated using an ExoQuick^™ Exosome Precipitation Kit (System Biosciences, LLC) according to the manufacturer's instructions. Briefly, plasma samples (500 µl) were combined with exosome precipitation reagent (100 µl) and incubated at room temperature for 10 min. The mixture was then subjected to centrifugation at a speed of 10,000 × g for a duration of 10 min at 4°C. The resulting pellet containing the exosomes was subsequently resuspended in PBS for subsequent analysis.

To isolate exosomes from the UCEC cell lines (HEC-1-A-Exo and Ishikawa-Exo), HEC-1-A or Ishikawa cells were cultured in RPMI-1640 medium supplemented with 10% FBS, at a temperature of 37°C at 5% CO₂. Following a culture period of 72 h, the supernatant containing exosomes and cellular debris was separated using centrifugation (10,000 × g for 30 min at 4°C). Exosomes from the collected supernatant were extracted using the GM^™ Exosome Isolation Reagent kit (Guangzhou Geneseed Biotech. Co., Ltd.). The resulting mixture was incubated at a temperature of 4°C for 30 min before centrifugation at 2,000 × g for 30 min. Finally, the isolated exosomes were PBS-resuspended for further analysis.

Exosome structure was analyzed using transmission electron microscopy (TEM; Hitachi, Ltd.) as previous described (23). Nanoparticle tracking analysis (NTA) was conducted using the NanoSight NS300 (Malvern Panalytical, Ltd.) to analyze the size distribution of the isolated exosomes. Western blotting was utilized to identify the presence of exosomal indicators, including CD9, CD81, tumor susceptibility gene 101 (TSG101) and calnexin.

The HEC-1-A donor cells (transfected with miR-32-5p-inhibitor-NC or miR-32-5p inhibitor) or Ishikawa donor cells (transfected with miR-32-5p-mimic-NC or miR-32-5p mimic) were seeded on a Transwell polyester permeable support. Simultaneously, the receptor HEC-1-A or Ishikawa cells (both untransfected) were cultured in the lower chamber of Transwell culture plate. After a 24 h incubation, the receptor cells were collected for subsequent experiments.

TRIzol (Invitrogen; Thermo Fisher Scientific, Inc.) was used for RNA isolation from UCEC cell lines according to the manufacturer's instructions. The Multi-type Sample DNA/RNA Extraction-Purification Kit (Sansure Biotech, Inc.) was used to isolate RNA from exosomes according to the manufacturer's instructions. The RNA concentration was measured using a Nanodrop spectrophotometer (Thermo Fisher Scientific, Inc.). The optical density 260/280 nm ratios of all RNA samples were ≥1.8. cDNA was obtained using the ReverTra Ace qPCR RT Kit (Toyobo Co., Ltd.) according to the manufacturer's instructions and subject to PCR analysis using a SYBR High-Sensitivity qPCR Supermix Kit (Novoprotein Scientific, Inc.) with the Real-time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used for the PCR: Initial denaturation at 95°C for 10 min; 40 cycles of 95°C for 15 sec and 60°C for 30 sec. The 2^−ΔΔCq values (24) reflected the RNA expression levels using U6 (for miR-32-5p) (25,26) and GAPDH (for AKT, PI3K, Bcl-2 and FOXN2) as controls. qPCR primers used are shown in Table II.

The viability of UCEC cells was measured via CCK-8 assay. HEC-1-A or Ishikawa cells (3×10³ cells/ml) were grown in 96-well plates for 0, 24, 48 and 72 h before adding the CCK-8 solution (15 µl; Dojindo Laboratories, Inc.). Following a 2 h incubation, UCEC cell viability was calculated via microplate reader (Bio-Rad Laboratories, Inc.), with the absorbance at a wavelength of 450 nm.

Colony forming experiments were carried out to further estimate the proliferative capacities. HEC-1-A or Ishikawa cells (~1,000 cells) were seeded into each well of 6-well plates and cultured for 2 weeks. The culture medium was refreshed biweekly. Following that, the cells were rinsed with PBS, treated with 100 % methanol for fixation at 25°C for 10 min and subsequently dyed with 0.1% crystal violet (Sigma-Adrich; Merck KGaA) for a duration of 20 min at 25°C. Cells were then washed three times, and cell colonies were counted using a light microscope (Olympus Corporation).

The 5-ethynyl-2′-deoxyuridine (EdU) proliferation assay Kit (Abcam) was used according to the manufacturer's instructions. HEC-1-A or Ishikawa cells were initially incubated with 50 µM EdU at 37°C for 2 h. Fixation was performed using 4% formaldehyde at 25°C for 15 min and permeabilized with 0.5% Triton X-100 at 25°C for 20 min. Following this, the cells were incubated with an Apollo reaction cocktail (1X; Abcam) at room temperature for ~30 min. To visualize DNA, cells were stained with DAPI at 25°C for 30 min. A fluorescence microscope (Carl Zeiss AG) was utilized to observe the EdU-positive cells.

The upper chamber of the Transwell insert was pre-coated with 50 µl of Matrigel at 37°C for 30 min, which had been diluted 5-fold in serum-free RPMI-1640. HEC-1-A or Ishikawa cells (5×10⁴ cells) were suspended in a RPMI-1640 medium without serum and then placed into the upper chamber. Simultaneously, the lower chamber was supplemented with RPMI-1640 medium containing 10% FBS. Following overnight incubation at 37°C, the cells in the lower chamber were treated with 0.1% crystal violet at 37°C for 15 min. Images of the stained cells were captured using a light microscope (magnification, ×400).

The apoptotic UCEC cells were assessed by utilizing the Annexin V-FITC apoptosis detection kit (Thermo Fisher Scientific, Inc.) following the guidelines provided by the manufacturer. The HEC-1-A or Ishikawa cells (2×10⁵) were suspended in 500 µl of binding buffer and then treated with Annexin V-EGFP and PI (5 µl each) at a temperature of 4°C for a duration of 15 min in the absence of light. Thereafter, the apoptosis of the cells was evaluated using a FACScan flow cytometer (Becton, Dickinson and Company) and analyzed using the BD CellQuest software (version 3.3; Becton, Dickinson and Company).

The mRNA targets of miR-32-5p were predicted using the StarBase software (version 2.0; http://starbase.sysu.edu.cn/), and 2,143 targets were predicted. FOXN2 was selected for the subsequent experiments due to its key role in endometrial cancer (27) and unknown regulatory relationship with miR-32-5p.

The 3′-untranslated region (UTR) of FOXN2 containing the putative binding sites of miR-32-5p was cloned into the luciferase reporter plasmid pGL3 vector (Promega Corporation) to construct the FOXN2 wild-type (WT)/mutant type (MUT). Ishikawa cells (5×10³) were seeded onto a 24-well plate and co-transfected with one of the aforementioned plasmids (80 ng), along with either miR-32-5p mimic or mimic-NC (10 nM), using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.). After 48 h at 37°C, the relative luciferase activity was determined using a Dual-Luciferase Reporter Assay System Kit (Promega Corporation). The activity of firefly luciferase was normalized to that of Renilla luciferase.

RIPA lysis buffer (Beyotime Institute of Biotechnology) containing protease inhibitors was utilized for protein extraction from cells, and protein concentrations were determined using the BCA Protein Assay Kit (Beyotime Institute of Biotechnology). In each lane, ~30 µg of proteins was separated using 10% SDS-PAGE and transferred onto a PVDF membrane. Blocking of the membrane was carried out at room temperature using 5% bovine serum albumin (Beyotime Institute of Biotechnology) for 2 h at 25°C. Afterward, the membrane was placed in a 4°C environment and incubated overnight with primary antibodies against AKT (1:1,500; cat. no. #9272; Cell Signaling Technology), p-AKT (1:1,500; cat. no. #4060; Cell Signaling Technology), PI3K (1:1,500; cat. no. #4249; Cell Signaling Technology), Bcl-2 (1:1,500; cat. no. #3498; Cell Signaling Technology), FOXN2 (1:1,500; cat. no. ab236385; Abcam), GAPDH (1:1,500; cat. no. #2118; Cell Signaling Technology), Calnexin (1:1,500; cat. no. #2679; Cell Signaling Technology), CD9 (1:1,500; cat. no. #13174; Cell Signaling Technology), CD81 (1:1,500; cat. no. ab79559; Abcam) and TSG101 (1:1,500; cat. no. ab125011; Abcam). Subsequently, the membranes were washed three times using tris-buffered saline with 0.05% Tween 20. Then, the HRP-conjugated secondary antibody (1:3,000; cat. no. #7074; Cell Signaling Technology) was added at room temperature for 1 h. The reference gene used was GAPDH. The immunoreactive protein bands were observed using an ECL Basic Kit (ABclonal Biotech Co., Ltd.) using a Gel-Pro analyzer (version 4.0; Media Cybernetics, Inc.).

Data analysis was conducted utilizing SPSS software (version 20.0; IBM Corp.). The findings were presented as the mean value ± standard deviation. Student's t-test (unpaired) or one-way ANOVA with Tukey's post hoc test was used for comparisons. Multivariate Cox regression analysis was performed on DE-miRNAs utilizing the Survival package in R (Posit Software, PBC). The median was selected as the cut-off value, and Kaplan-Meier survival curves were generated using the Survminer package (version 0.5.0; DataNovia) and analyzed using log-rank test. P<0.05 was considered to indicate a statistically significant difference.

The present study screened miRNA expression data from patients with UCEC of TCGA database and DE-miRNA data from Zhou et al (21). Compared with that in normal samples, there were 656 miRNAs differentially expressed in UCEC tissues from TCGA database (Fig. 1A). Among them, 402 miRNAs were differentially upregulated, while 254 miRNAs were differentially downregulated in UCEC. The heatmap of top 50 DE-miRNAs is shown in Fig. S1. Through intersecting these 656 DE-miRNAs with DE-miRNA data (comprising 49 DE-miRNAs) from Zhou et al (21), a total of 26 DE-miRNAs were obtained (Fig. 1B). Univariate Cox risk regression analysis (P<0.05) was conducted on the TCGA data, and 140 miRNAs were identified as potential prognostic miRNAs in patients with UCEC. These 140 miRNAs were then intersected with 26 DE-miRNAs from the aforementioned analysis, from which 7 overlapping DE-miRNAs were identified (Fig. 1C). Multivariate Cox regression analysis was used to demonstrate that miR-15a-5p (P=0.013), miR-1180-3p (P=0.014), miR-32-5p (P=0.002) and miR-1197 (P=0.037) were independent prognostic factors of UCEC (Fig. 1D). Survival probability analysis was conducted to investigate the effects of the expression of these four miRNAs on survival outcomes, from which miR-32-5p demonstrated the most significant effects on survival probability and was therefore selected for further investigation (P=0.00036; Fig. 1E).

To confirm the successful isolation of exosomes from plasma samples from patients with UCEC and healthy controls, TEM and NTA were employed to characterize the isolated exosomes. The isolated vesicles exhibited a double-layered membrane structure and a size distribution of ~100 nm in diameter, which were in accordance with the expected characteristics of exosomes (Fig. 2A and B) (28). Western blotting demonstrated that the vesicles expressed the exosome-positive markers CD9, CD81 and TSG101, while the exosome-negative marker, Calnexin, was absent in the isolated vesicles (Fig. 2C), which demonstrated successful isolation of exosomes. The expression of miR-32-5p in normal-Exo and UCEC-Exo was then determined; miR-32-5p expression was upregulated in UCEC-Exo compared with that of normal-Exo (P<0.05; Fig. 2D).

Subsequently, exosomes were isolated from two UCEC cell lines (HEC-1-A and Ishikawa). The characteristics of vesicles from UCEC cell lines were similar to exosomes isolated from plasma samples (Fig. 3A and B). Western blotting demonstrated presence of exosome-positive markers CD9, CD81 and TSG101 in the exosomes-enriched fractions, and absence of Calnexin (Fig. 3C). Additionally, compared with that of HEC-1-A-Exo, miR-32-5p expression was downregulated in Ishikawa-Exo (P<0.001; Fig. 3D).

miR-32-5p expression levels were significantly reduced in Ishikawa cells compared with that of HEC-1-A cells (P<0.001; Fig. 4A). As miR-32-5p demonstrated high expression levels in HEC-1-A cells, they were selected for transfection with the miR-32-5p inhibitor to achieve interference in miR-32-5p expression, whereas the Ishikawa cells were chosen for transfection with the miR-32-5p mimic to achieve overexpression, due to the low expression levels of miR-32-5p in Ishikawa cells. qRT-PCR results that miR-32-5p expression in HEC-1-A cells was significantly decreased following transfection with the miR-32-5p inhibitor (P<0.001), while miR-32-5p was overexpressed in Ishikawa cells upon transfection with the miR-32-5p mimic (Fig. 4B; P<0.001). miR-32-5p inhibition significantly suppressed the viability of HEC-1-A cells, as evidenced by a CCK-8 assay (Fig. 4C; P<0.001). Conversely, overexpression of miR-32-5p increased the cell viability of Ishikawa cells (Fig. 4C; P<0.001). Colony formation and EdU proliferation assays further confirmed the impacts of miR-32-5p expression status on cell proliferation. Transfection with miR-32-5p inhibitor led to a significant decrease in colony formation efficiency and the percentages of EdU-positive cells in HEC-1-A cells (Fig. 4D and E; P<0.05), while transfection with miR-32-5p mimic significantly increased colony formation efficiency and EdU-positive cell percentages in Ishikawa cells (P<0.01). The migratory abilities of UCEC cell were assessed using a Transwell assay, demonstrating that downregulation of miR-32-5p significantly suppressed HEC-1-A cell migration (Fig. 4F; P<0.01), whereas miR-32-5p overexpression significantly increased the migratory capacities of Ishikawa cells (P<0.01). Furthermore, flow cytometry analysis demonstrated increased apoptosis in HEC-1-A cells upon treatment with miR-32-5p inhibitor (Fig. 4G; P<0.001), while Ishikawa cells exhibited a significantly reduced apoptotic rate following treatment with miR-32-5p mimic (P<0.001).

The Starbase software predicted a potential binding site between miR-32-5p and FOXN2 (Fig. 5A). The DLR assay was conducted in Ishikawa cells, which demonstrated a significant decrease in relative luciferase activity in the FOXN2-WT + miR-32-5p mimic group when compared with that of the FOXN2-WT + mimic-NC group (Fig. 5B; P<0.001), while no statistical changes were observed between the FOXN2-MUT + miR-32-5p mimic and FOXN2-MUT + mimic-NC groups. These findings suggested that FOXN2 was a direct target of miR-32-5p. The PI3K/AKT pathway is a well-established mechanism implicated in the promotion of cancer development (29,30), including in UCEC (31), but whether it is regulated by miR-32-5p in UCEC remains unknown. Therefore, we further explored the effects of miR-32-5p on PI3K/AKT pathway in UCEC cells. Inhibition of miR-32-5p resulted in a significant inhibition of AKT and PI3K mRNA expression (Fig. 5C; P<0.001). Additionally, significantly decreased Bcl-2 mRNA expression levels (P<0.05) and increased FOXN2 mRNA expression levels (P<0.01) were observed upon miR-32-5p inhibition. The opposite effects were observed in the results of mRNA expression of AKT, PI3K, Bcl-2 and FOXN2 in Ishikawa cells following transfection of miR-32-5p mimic (Fig. 5D; P<0.05). Western blotting was used to assess protein levels in HEC-1-A and Ishikawa cells. The expression levels of AKT protein were unaltered by both miR-32-5p overexpression and inhibition (Fig. 5E and F). When transfected with miR-32-5p inhibitor in HEC-1-A cells, the protein expression levels of p-AKT, PI3K and Bcl-2 were significantly decreased (P<0.05), but FOXN2 protein expression levels were increased (P<0.001). The ratio of p-AKT/AKT was significantly decreased in HEC-1-A cells transfected with miR-32-5p-inhibitor (P<0.001). However, overexpression with miR-32-5p mimic showed the opposite effects (P<0.05). These findings suggested that miR-32-5p negatively regulated FOXN2 and activated PI3K/AKT/Bcl-2 pathway. Rescue experiments were performed in Ishikawa cells to further verify the effects of miR-32-5p-mediated FOXN2 on PI3K/AKT/Bcl-2 pathway. mRNA expression of FOXN2 was significantly increased followed transfection of OE-FOXN2 (Fig. 6A; P<0.0001), which demonstrated the successful transfection. Overexpression of FOXN2 significantly reversed the effects of miR-32-5p-mimic transfection that promoted the activation of PI3K/AKT/Bcl-2 pathway (Fig. 6B and C; P<0.05), and the inhibitory effect on FOXN2 expression levels (Fig. 6B and C; P<0.001).

The impacts of Exo-miR-32-5p on the proliferation, migration and apoptosis of UCEC cells were examined utilizing a co-culture model. Firstly, Exo-miR-32-5p expression in HEC-1-A cells transfected with miR-32-5p-inhibitor or in Ishikawa cells transfected with miR-32-5p-mimic was determined using qRT-PCR. miR-32-5p expression was significantly downregulated in Exo-miR-32-5p-inhibitor group when compared with the Exo-miR-32-5p-inhibitor-NC group (Fig. 7A; P<0.001), while miR-32-5p expression levels were significantly increased in the Exo-miR-32-5p-mimic group when compared with the Exo-miR-32-5p-mimic-NC group (Fig. 7A; P<0.001). Compared with the Exo-miR-32-5p-inhibitor-NC group, the proliferation and migration of HEC-1-A cells were significantly decreased in the Exo-miR-32-5p-inhibitor group (Fig. 7B-E; P<0.01), while the rate of apoptosis was increased (Fig. 7F; P<0.001). Conversely, compared with the Exo-miR-32-5p-mimic-NC group, a significant increase of proliferative and migratory capacities (Fig. 7B-E; P<0.01) and significant decrease of the apoptosis rate in Ishikawa cells of the Exo-miR-32-5p-mimic group was demonstrated (Fig. 7F; P<0.001). Additionally, FOXN2 levels were increased, Bcl-2 levels were decreased and the PI3K/AKT pathway was inactivated in HEC-1-A cells (Fig. 7G and H; P<0.01), while these effects were all reversed in Ishikawa cells (Fig. 7G and H; P<0.01).