Medical records from the National Cancer Center Hospital (Tokyo, Japan) were retrospectively reviewed. Excluding patients without written informed consent, all six patients with ULMS who underwent surgery without neoadjuvant therapy between January 2011 and September 2020 were included. The archival fresh-frozen tumor and adjacent normal tissues of these patients, which were stored at the National Cancer Center Biobank (Tokyo, Japan), were used in the present study. Moreover, three leiomyoma tissues from three other patients were used as controls. The case number corresponds to the case number from our previous report (13). The clinical information, such as age, stage, mitotic rate and the presence of necrosis, was obtained from their clinical records. The International Federation of Gynecology and Obstetrics staging system was used (2,3). The study protocol was approved by the ethics committee of the National Cancer Center (approval no. 2020-160). Written informed consent was obtained from all patients. Moreover, the study was carried out according to The Declaration of Helsinki.

Total RNA was extracted using the miRNeasy Mini Kit (Qiagen GmbH), and small RNA libraries were prepared using the NEBNext Multiplex Small RNA Library Prep Set for Illumina (cat. no. E7300L; New England Biolabs, Inc.) according to the manufacturers' protocol. Subsequently, the small RNA libraries were separated by electrophoresis (120 V, 60 min) on a 10% TBE gel (cat. no. EC6265BOX; Thermo Fisher Scientific, Inc.), and DNA fragments corresponding to 140–160 bp (the lengths of small non-coding RNA plus the 3′ and 5′ adaptors) were recovered. The cDNA concentration was then measured using the Qubit dsDNA HS Assay Kit and a Qubit2.0 Fluorometer (Thermo Fisher Scientific, Inc.). Finally, single-end reads were performed using the MiSeq Reagent Kit v3 (cat. no. MS-102-3001; Illumina, Inc.) on the Illumina MiSeq (Illumina, Inc.) and the loading concentration of the final library was 10 pM.

The CLC Genomics Workbench version 9.5.3 program (Qiagen GmbH) was used for adaptor trimming and mapping to the miRbase 21 database (https://www.mirbase.org/) without allowing any mismatch. After normalization using reads per million mapped reads, low-expressed miRNAs (<10 reads in all samples) were excluded from further analyses. Subsequently, RStudio (RStudio, Inc.) and R software (version 4.0.3; http://www.r-project.org/) were used. For the heatmap analysis, miRNAs with an absolute log2 (fold change) of >0.8 were extracted and utilized. The data were then converted to base 10 logarithms and z-scores, and the heatmap.2 function of the gplots package (ver. 3.1.0; http://cran.r-project.org/package=gplots) was used. The P-values for each gene were calculated using the Wald test in DESeq2 (ver. 1.30.0) for the volcano plots (22).

SK-UT-1 (ULMS-derived cell line) and SK-LMS-1 (vulvar LMS-derived cell line) were purchased from the American Type Culture Collection. The cells were maintained in MEM (Nacalai Tesque, Inc.) containing 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Inc.), 1 mM sodium pyruvate (Thermo Fisher Scientific, Inc.) and penicillin-streptomycin (Thermo Fisher Scientific, Inc.) in a humidified incubator at 37°C with 5% CO₂. The cell lines tested negative for mycoplasma contamination and were used between 5 and 40 passages for the experiments.

mirVana miRNA mimics (Thermo Fisher Scientific, Inc.) were used to induce overexpression of miRNAs in the present study, and the assay IDs were as follows: miR-10b-5p (MC11108), miR-29a-3p (MC12499), miR-126-3p (MC12841), miR-186-5p (MC11753) and Negative Control (NC) #1 (4464058). Cells were transfected with 20 nM miRNA mimics using Lipofectamine^® RNAi Max (Thermo Fisher Scientific, Inc.) at 37°C for ≥24 h.

Total RNA was extracted from clinical samples or cell lines as aforementioned, and cDNA was synthesized using a TaqMan Advanced miRNA cDNA Synthesis Kit (Thermo Fisher Scientific, Inc.) according to the manufacturers' protocol. Subsequently, qPCR was performed using TaqMan Fast Advanced Master Mix and TaqMan Advanced miRNA Assay (Thermo Fisher Scientific, Inc.); the assay IDs were as follows: miR-10b-5p (478494_miR), miR-29a-3p (478587_miR), miR-126-3p (477887_miR), miR-186-5p (477940_miR) and RNU6B (001093). The amplification program was as follows: Denaturation at 95°C for 10 min, followed by 40 amplification cycles at 95°C for 15 sec and 60°C for 60 sec. The amplified product was monitored by measuring the fluorescence intensity of FAM. U6 was used as a reference gene to normalize the expression and the 2^−ΔΔCq method was used for quantification (23).

Cells were seeded into 96-well plates (1,000 cells/well) and transfected with the miR-10b-5p, miR-29a-3p, miR-126-3p, miR-186-5p or NC mimics. A total of 24, 48 and 72 h post-transfection, cell proliferation was assessed using the CellTiter-Glo 2.0 Cell Viability Assay (Promega Corporation), and luminescence was measured 10 min after adding the reagent using SpectraMax iD3 (Molecular Devices, LLC) or Infinite 200 PRO (Tecan Group, Ltd.). Experiments were performed in triplicate and repeated three times.

Cells were transfected with miR-10b-5p or NC mimic in 35-mm dishes (50,000 cells/dish). A total of 24 h post-transfection, the cells were seeded into six-well plates (300 cells/well, six replicates) and incubated for 6 days in a humidified incubator at 37°C with 5% CO₂. Subsequently, the cells were fixed with 4% paraformaldehyde for 10 min and stained with 1% crystal violet for 10 min at room temperature, and the colonies (>50 cells) were manually counted.

For soft agar colony formation assay, 2X MEM was prepared using 10X MEM (cat. no. M0275; MilliporeSigma), FBS, sodium hydrogen carbonate (FUJIFILM Wako Pure Chemical Corporation), GlutaMax (Thermo Fisher Scientific. Inc.), sodium pyruvate and penicillin-streptomycin (Thermo Fisher Scientific, Inc.). Cells were transfected with miR-10b-5p or NC mimic in 35-mm dishes (50,000 cells/dish). Subsequently, 1.6 and 0.6% agar solutions were prepared using agar powder (FUJIFILM Wako Pure Chemical Corporation) diluted with PBS. Prewarmed 2X MEM and melted 1.6% agar solution were mixed (1:1 ratio) and transferred into six-well plates to form the bottom agar layer. Then, a total of 24 h post-transfection, the cells were trypsinized and resuspended in a prewarmed 2X MEM. The cell suspension and melted 0.6% agar solution were mixed (1:1 ratio) and placed on the bottom agar layer (3,000 cells/well). The cells were incubated with culture medium for 14 days in a humidified incubator at 37°C with 5% CO₂. Then, cells were stained with 0.01% crystal violet for 1 h at room temperature, and the colonies (>50 cells) were manually counted. Images were captured using a WRAYCAM-NF300 light microscope (WRAYMER Inc.).

Cells were transfected with miR-10b-5p or NC mimic in six-well plates (50,000 cells/well). A total of 48 h post-transfection, the cells were harvested trypsinization and washed with 3% FBS in PBS. Then, the cells were fixed in cold 70% ethanol with gentle vortexing and were placed in 70% ethanol at −20°C for 24 h. The cells were centrifuged at 500 × g for 15 min at 20°C and resuspended in 3% FBS in PBS. After centrifuging, the cell pellet was stained with 0.5 ml PI/RNase Staining Buffer (BD Biosciences) for 15 min at room temperature. The FACSCanto II flow cytometer (BD Biosciences) was used for cell cycle analysis. The resulting data were analyzed using FlowJo software (version 10.8.1; FlowJo LLC). Experiments were performed in triplicate.

Cells were transfected with miR-10b-5p or NC mimic in six-well plates (50,000 cells/dish). A total of 48 h post-transfection, total RNA was extracted as aforementioned. Pair-end sequencing was performed by Azenta Life Sciences. Briefly, total RNA was quantified and qualified using the Qubit RNA HS Assay Kit (Thermo Fisher Scientific, Inc.) and TapeStation RNA ScreenTape (Agilent Technologies, Inc.). To enrich poly-A mRNA and to remove rRNA molecules, the NEBNext Poly(A) mRNA Magnetic Isolation Module (cat. no. E7490L; New England Biolabs, Inc.) was used. Subsequently, cDNA synthesis followed by transcriptome library preparation was conducted using the MGIEasy RNA Directional Library Prep Kit V2.0 (cat. no. 1000005272; MGI Tech Co., Ltd.). The resulting sequencing libraries were quantified using the Qubit DNA HS Assay Kit (Thermo Fisher Scientific, Inc.) and their fragment size distribution was confirmed by TapeStation D1000 ScreenTape (Agilent Technologies, Inc.). The double-stranded library fragments were pooled/multiplexed at an equimolar amount and further processed into single-stranded circular DNA (sscDNA). The sscDNA libraries were quantified using the Qubit ssDNA Assay Kit (Thermo Fisher Scientific), and a 40 fmol sscDNA library pool was used for generating DNA nanoballs (DNBs) by rolling circle replication reaction. DNBs were then loaded into a flow cell for sequencing on the DNBSEQ-G400 platform (MGI Tech Co., Ltd.) with 150 bp paired-end configuration, according to the manufacturer's instructions. From the sequencing data, expression levels for each gene were quantified by Kallisto (ver. 0.46.2) (24). Then, data were summarized using the tximport package (ver. 1.18.0) of R software, and scaled transcript per million counts were used for further analyses (25). Genes with low read coverage (maximum read count, <10 reads) were excluded. Compared with NC-transfected cells, genes with absolute log2 (fold change) >0.8 were considered differentially expressed genes (DEGs). Subsequently, common DEGs in both cell lines were used to generate the heatmap after converting the data to base 10 logarithms and z-scores. The heatmap.2 function of the gplots package (ver. 3.1.0; http://cran.r-project.org/package=gplots) was used. Pathway analysis was performed using the Ingenuity Pathway Analysis (IPA) software (ver. 84978992; Qiagen GmbH).

Data are presented as the mean ± standard error of the mean and experiments were performed at least in triplicate and repeated three times. All statistical analyses were performed using RStudio and R software (ver. 4.0.3). Welch's t-test was used to determine the significant differences between the means of two sets of data, and the paired t-test was used to compare the expression of miR-10b-5p in paired ULMS and adjacent normal tissues. P<0.05 was considered to indicate a statistically significant difference.

Small RNA sequencing was performed using archival fresh-frozen samples from six patients with ULMS and three patients with myoma. Table I shows the clinical information of the patients. The heatmap shown in Fig. 1A indicated that the miRNA profiles of patients with ULMS were usually different from those of patients with myoma. However, the miRNA profiles were diverse in patients with ULMS; notably, the miRNA expression pattern in ULMS-3 was more similar to that of myoma compared with the other types of ULMS. Our previous study reported that ULMS-3 was characterized by higher ESR1 expression and a lower mitotic rate than other types of ULMS, suggesting that ULMS-3 is a gynecological subtype and a clinically less aggressive subtype of LMS (13,26,27). Subsequently, a volcano plot that compares ULMS to myoma was generated to investigate ULMS-associated miRNAs. The volcano plot revealed that 53 and 11 miRNAs were significantly upregulated or downregulated, respectively, in ULMS compared with in myoma (Fig. 1B). The normalized read counts of the 64 miRNAs are shown in Table SI. miRNAs with abundant expression were selected according to the baseline expression level. Dot plots of the top four downregulated miRNAs (miR-10b-5p, miR-29a-3p, miR-126-3p and miR-186-5p) and the top four upregulated miRNAs (miR-10a-5p, miR-146a-5p, miR-181a-5p and miR-181b-5p) are shown in Fig. 1C. In particular, the mean normalized read count of miR-10b-5p was 93,650 reads in myoma; however, it was markedly decreased to 27,903 reads in ULMS. Thus, miR-10b-5p was considered to serve a role in ULMS progression. RT-qPCR was performed to validate miR-10b-5p downregulation in ULMS; it was revealed that miR-10b-5p expression was significantly downregulated in ULMS tissues compared with that in paired normal tissues (P<0.01; Fig. 1D).

A gain-of-function analysis was performed to elucidate the potential roles of the downregulated miRNAs in LMS. The expression levels of miR-10b-5p were significantly increased post-transfection with the miR-10b-5p mimic (Fig. 2A). The overexpression of miR-10b-5p significantly decreased the proliferation of SK-UT-1 and SK-LMS-1 cell (P<0.01 and P<0.05; Fig. 2B). Similarly, miR-29a-3p, miR-126-3p and miR186-5p had tumor-suppressive roles; in particular, miR-126-3p significantly decreased the proliferation of SK-UT-1 and SK-LMS-1 cells (P<0.01 and P<0.001; Fig. S1A-F). The present study focused on miR-10b-5p because the baseline expression of miR-10b-5p was ~10-fold higher than that of miR-126-5p (Fig. 1C). Subsequently, the clonogenic assay revealed that miR-10b-5p overexpression significantly reduced the number of SK-UT-1 and SK-LMS-1 cell colonies (P<0.001 and P<0.001; Fig. 2C). However, soft agar colony formation assay showed that miR-10b-5p did not suppress the number of colonies (Fig. S2). In addition, the overexpression of miR-10b-5p significantly increased the population of SK-UT-1 and SK-LMS-1 cells in G₁ phase (P<0.05 and P<0.01), and decreased the population of SK-UT-1 and SK-LMS-1 cells in G₂/M phase (SK-UT-1 and SK-LMS-1; P<0.05 and P<0.01) (Fig. 2D). Therefore, these results may suggest that miR-10b-5p suppressed the proliferation of LMS cells.

Transcriptome analysis was performed to investigate the molecular background of miR-10b-5p-associated tumor suppression. miR-10b-5p-transfected SK-UT-1 cells had 1,830 upregulated and 628 downregulated genes compared with NC-transfected SK-UT-1 cells. Similarly, 812 upregulated and 1,133 downregulated genes were observed in miR-10b-5p-transfected SK-LMS-1 cells. Of these, 300 upregulated and 214 downregulated genes were identified in both cell lines (Fig. 3A). The heatmap in Fig. 3B shows the levels of 514 common DEGs in these cells. Subsequently, IPA analysis was performed using the 514 DEGs and 52 significantly dysregulated pathways were revealed (Fig. 3C; Table SII). For example, ‘Cell Cycle: G1/S Checkpoint Regulation’ (P=3.31×10⁻⁴; z-score=−0.378) and ‘MYC Mediated Apoptosis Signaling’ (P=3.80×10⁻⁴; z-score=−0.816) were significantly inhibited, whereas ‘Regulation Of The Epithelial-Mesenchymal Transition In Development Pathway’ (P=1.45×10⁻³; z-score, 2.000) was significantly activated (Fig. 3C).