All the subjects were recruited from Hainan Branch of PLA General Hospital (Sanya, China) between January 2015 and July 2016. Written informed consent was obtained prior to inclusion and the present study was approved by the Ethics Committee of Hainan branch of PLA General Hospital. A total of 30 endometrial tissues were collected from IUA patients diagnosed by hysteroscopy, with mean age 28.06 years (range 19–34). Normal endometrial tissues from individuals without IUA (n=15) in the same period were used as controls. All samples were flash-frozen in liquid nitrogen, and stored at −80°C until subsequent molecular analysis.

Total RNA was extracted from endometrial tissues using the miRcute miRNA Isolation kit (Tiangen Biotech Co., Ltd., Beijing, China). RNA quality was determined using an Agilent Bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA, USA). Samples were labeled using a miRCURY Hy3^™/Hy5^™ Power labeling kit and hybridized to the miRCURY LNA^™ Array version 18.0 (Agilent Technologies, Inc.). The chips were scanned with the Axon Gene Pix 4000B Microarray Scanner (Axon Instruments, Foster City, CA, USA). The procedure and image processing method was performed as previously described (24). The miRNA expressions of all differentially expressed samples were clearly displayed by a hierarchical clustering heat map.

Total RNA was isolated from endometrial tissues using the TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocol. For the detection of miR-326, RT-qPCR assays were performed using the miRcute miRNA qPCR Detection kit (Tiangen Biotech Co., Ltd.) following the manufacturer's protocol. For detection of the TGF-β1, α-smooth muscle actin (α-SMA), collagen type I α 1 chain (COL1A1) and fibronectin (FN) mRNA levels, 1 µg of total RNA was reverse transcribed at 37°C for 60 min using the miRcute miRNA First-Strand cDNA Synthesis kit (Tiangen Biotech Co., Ltd.), then amplified on an Applied Biosystems 7900HT cycler using SuperReal PreMix Plus (Tiangen Biotech Co., Ltd.). The PCR conditions consisted of 5 min at 94°C for one cycle followed by 40 cycles of 94°C for 20 sec and 60°C for 20 sec. U6 and GAPDH were used as normalization controls for miRNA qPCR and qPCR, respectively. The relative expression was calculated using the 2^−ΔΔCq method (25). Each reaction was conducted in triplicate. The primers for RT-qPCR analysis were as follows: miR-326 forward primer: 5′-GGCGCCCAGAUAAUGCG-3′; miR-326 reverse primer: 5′-CGTGCAGGGTCCGAGGTC-3′; U6 forward primer: 5′-TGCGGGTGCTCGCTTCGCAGC-3′; U6 reverse primer: 5′-CCAGTGCAGGGTCCGAGGT-3′; TGF-β1 forward primer: 5′-TGGACCGCAACAACGCCATCTATGAGAAAACC-3′, reverse primer: 5′-TGGAGCTGAAGCAATAGTTGGTATCCAGGGCT-3; COL1A1 forward primer: 5′-GAGGGCCAAGACGAAGACATC-3, reverse primer: 5′-CAGATCACGTCATCGCACAAC-3; α-SMA forward primer: 5′-GGCTCTGGGCTCTGTAAGG-3, reverse primer: 5′-CTCTTGCTCTGGGCTTCATC-3; fibronectin forward primer: 5′-GAGAGATCTGGAGGTCAT-3′, reverse primer: 5′-GGGTGACACCTGAGTTGAA-3′; GAPDH reverse primer: 5′-CGGAGTCAACGGATTTGGTCGTAT-3′, reverse primer: 5′-AGCCTTCTCCATGGTGGTGAAGAC-3′.

ESCs were isolated from the endometrial tissues as previously described (26). ESCs were maintained in Dulbecco's modified Eagle's medium/Nutrient Mixture F-12 (DMEM/F12; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin, 100 µg/ml streptomycin at 37°C in a humidified atmosphere of 5% CO₂. The cells reached confluence in 2–3 days, and the passage 3–6 were used for subsequent experiments.

MiR-326 mimics, miR-326 inhibitor and controls were purchased from Shanghai GenePharma Co., Ltd., (Shanghai, China). ESCs were seeded at 2×10⁵ cells/well in 6-well plates and cultured in antibiotic-free DMEM at 37°C and 5% CO₂. Next, the cells were incubated overnight to a confluence of 30–50%, cell transfection was performed using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Cells were collected for further experiments 48 h after the transfection.

A cDNA fragment of the TGF-β1 3′-UTR mRNA containing the seed sequence of the miR-326-binding site or a mutated binding site was cloned into the pmirGLO dual-luciferase vector (Promega Corporation, Madison, WI, USA). The constructed dual-luciferase vector was co-transfected with miR-326 mimics, miR-326 inhibitor or NC into ESCs isolated from patients with IUA using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The relative firefly luciferase activity normalized with Renilla luciferase was measured 48 h after transfection by using the Dual-Luciferase Reporter system (Promega Corporation) following the manufacturer's protocol. All of the dual-luciferase reporter assays were performed in triplicate for each experiment and three independent experiments were conducted.

Cells were lysed using radioimmunoprecipitation assay protein extraction reagent (Beyotime Institute of Biotechnology, Haimen, China). The concentration of proteins was determined using a bicinchoninic acid assay kit (Pierce Biotechnology, Rockford, IL, USA). Protein extracts (50 µg per lane) were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, then transferred to nitrocellulose membranes (Sigma-Aldrich, Merck Millipore; Darmstadt, Germany) and incubated with the following primary antibodies: TGF-β1 monoclonal antibody (cat. no. sc-52892; 1:300; Santa Cruz Biotechnology, Inc., Dallas, TX, USA), anti-phosphorylated (p)-Smad3 monoclonal antibody (cat. no. sc-517575; 1:1,000; Santa Cruz Biotechnology, Inc.) and Smad3 monoclonal antibody (cat. no. sc-101154; 1:1,000; Santa Cruz Biotechnology, Inc.). Subsequently, the membranes were incubated with peroxidase conjugated goat anti-rabbit IgG (cat. no. TA130023; 1:5,000; OriGene Technologies, Inc., Beijing, China) conjugated to horseradish peroxidase for 1 h at 37°C. ECL chromogenic substrate (Thermo Fisher Scientific, Inc.) was used to visualize the bands and the intensity of the bands was quantified by densitometry using ImageJ software (version 2.1.4.7; National Institutes of Health, Bethesda, MD, USA). Control antibody used was anti-β-actin (cat. no. sc-8432; 1:1,000; Santa Cruz Biotechnology, Inc.).

Statistical analysis was performed using GraphPad Prism version 5.0 (GraphPad Software, Inc., San Diego, CA, USA). Differences were analyzed with Student's t-test between two groups or with one-way analysis of variance among four groups. Correlation analyses were performed using Spearman's correlation coefficient test. P<0.05 was considered to indicate a statistically significant difference.

To determine the expression of miRNAs associated with IUA, the present study performed a miRNA microarray on endometrial tissues from pairs of patients with or without IUA. The current findings revealed that the expression of 56 miRNAs was significantly different in the IUA group compared with the normal group and from these miRNAs, 28 miRNAs were downregulated, whereas 28 miRNAs were upregulated (Fig. 1A). From these aberrantly expressed miRNAs, miR-326 was one which was significantly downregulated. In addition, a previous study reported that miR-326 had a key role in alleviating lung fibrosis by regulating TGF-β1 expression and other pro-fibrotic genes (27). Therefore, the present study selected miR-326 for further investigation.

To validate the expression trend identified for miR-326 in endometrial tissues obtained from miRNA microarray assay RT-qPCR was performed to detect miR-326 expression levels in 30 IUA endometrial tissues and 15 normal endometrial tissues. As presented in Fig. 1B, miR-326 was significantly downregulated in the IUA group compared with the normal group. These findings indicated that miR-326 may be involved in IUA progression.

The present study examined whether abnormal expression of miR-326 was associated with the endometrial fibrosis of patients with IUA following confirmation of the reduced expression of miR-326 in endometrial tissues. Previous studies revealed that TGF-β1 may promote the endometrial fibrosis, and α-SMA, COL1A1 and FN are the primary fibrotic marker genes (9,20,28). Thus, the miR-326 expression level and the fibrotic markers (TGF-β1 α-SMA, COL1A1, and FN) mRNA expression level in 30 IUA endometrial tissues were determined. Using Spearman's correlation coefficient, a negative correlation was identified between TGF-β1, α-SMA, COL1A1, and FN mRNA expression levels and the miR-326 expression level (Fig. 2). These findings indicate that miR-326 may potentially serve as an effective biomarker for the prognosis of patients with IUA. Additionally, miR-326 may be relevant to fibrotic processes in IUA pathology.

Previous studies have reported that deregulated proliferation and differentiation of ESCs are important factors contributing to the development of endometrial fibrosis (29,30). Therefore, the present study used the primary ESCs isolated from patients with IUA to investigate function of miR-326 in endometrial fibrosis. ESCs were transfected with miR-326 mimics to overexpress miR-326. The miR-326 mimic group had a significantly increased the expression level of miR-326 (Fig. 3A). The alteration of the expression of fibrotic-associated markers in ESCs were investigated following miR-326 overexpression. As presented in Fig. 3B-D, the mRNA expression levels of α-SMA, COL1A1 and FN in ESCs isolated from patients with IUA was increased when compared with normal subjects, whereas overexpression of miR-326 was able to suppress the mRNA expression levels of α-SMA, COL1A1 and FN. These findings suggest that miR-326 may inhibit the fibrosis of endometrial stromal cells.

TGF-β1, an important pro-fibrogenic mediator (31), was previously reported to be involved in the development of fibrosis and organ dysfunction, by upregulating the extracellular matrix (ECM) protein expression and α-SMA (8). As miR-326 regulates the TGF-β1 level and attenuates bleomycin-induced lung fibrosis mice (27), it was hypothesized that TGF-β1 may also be involved in the anti-fibrotic role of miR-326 in IUA. In order to investigate the association between miR-326 and TGF-β1, a luciferase reporter assay was performed in ESCs isolated from patients with IUA. The findings revealed that overexpression of miR-326 significantly reduced the luciferase activity of TGF-β1 with wild-type (wt) 3′-UTR, whereas miR-326 inhibition increased the luciferase activity of TGF-β1 with wt 3′-UTR (Fig. 4A). Cells co-transfected with miR-326 mimics, miR-326 inhibitor, and TGF-β1-mut-3′UTR showed no obvious change in their luciferase activity (Fig. 4A). These findings indicate that miR-326 directly targeted the 3′-UTR of TGF-β1. The present study further examined whether miR-326 may modulate the expression of TGF-β1 in ESCs by RT-qPCR and western blotting. The findings revealed that the mRNA and protein expression levels of TGF-β1 in ESCs were significantly downregulated following transfection with miR-326 mimics, compared with the negative control, whereas transfection with the miR-326 inhibitor increased TGF-β1 expression (Fig. 4B and C). Collectively, these findings suggest that TGF-β1 is a direct downstream target of miR-326 in ESCs.

Due to the important role of the TGF-β1/Smad3 pathway in organ fibrosis, including endometrial fibrosis (23,32,33), subsequent experiments were designed to investigate the effects of miR-326 on the activity of the TGF-β1/Smad3 pathway in ESCs. Western blotting was performed to determine the protein expression levels of TGF-β1, p-Smad3 and Smad3 in ESCs. The present study confirmed that TGF-β1 and p-Smad3 protein expression levels were increased in ESCs from patients with IUA (Fig. 5A and B). However, overexpression of miR-326 effectively downregulated TGF-β1 and p-Smad3 protein expression levels (Fig. 5A and B). In addition, no change was observed in the expression levels of Smad3. These findings suggest that miR-326 may inhibit endometrial fibrosis via inactivating the TGF-β1/Smad3 pathway.