Clinical tissues were obtained from NSCLC patients who received surgical resection in Department of Oncology from 2002 to 2011. All these clinical tissues from patients have been pathologically diagnosed as NSCLC before performing experiments in this study. The informed consents were obtained from every patient involved in this study. Approval for experiments involving patient samples was obtained from the institutional research ethics committee of China-Japan Union Hospital of Jilin University (JL20010923).

Cell lines including A549, H1299, 95D, SPC-A1, H358 and BEAS-2B were from the Cell Bank of Chinese Academy of Sciences (Shanghai, China) and American Type Culture Collection (ATCC, Rockville, MD, USA). All cells were cultured in DMEM medium (Gibco Co., New York, NY, USA) supplemented with 10% fetal bovine serum (Gibco Co.). Cell cultures were incubated in cell incubators with humidified atmosphere and 5% CO₂ at 37°C.

Rnaeasy mini kit (Qiagen) were used to extract the RNA from NSCLC tissues and cells. Transcriptional First Strand cDNA Synthesis kit (Roche, Indianapolis, IN, USA) and SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA USA) were used for reverse transcription reactions and real-time PCR. Primers for miR-381, U6, LRH-1 and GAPDH were bought from Genecopoeia (Guangzhou, China). U6 and GAPDH were the internal controls for miR-381 and LRH-1, respectively.

Total protein was extracted in RIPA buffer (Beyotime, Shanghai, China) containing a protease and phosphatase inhibitor (Thermo Fisher Scientific, Rockford, IL, USA), then the protein concentration was determined using BCA Protein assay kit (Thermo Fisher Scientific). Equal protein (30 µg) was loaded and separated by 10% SDS-PAGE and transferred to polyvinylidenediflouride (PVDF) membrane (Millipore, Billerica, MA, USA). The membranes were incubated with respective primary antibodies: LRH-1 and GAPDH (1:1000, Cell Signaling Technology, Inc.) at 4°C overnight after blocking with 5% non-fat milk in TBS. Then the membranes were washed three times by TBST and incubated with appropriate peroxidase-conjugated secondary antibody for 2 h at room temperature (ZSGB-BIO, Beijing, China). Protein bands were visualized by enhanced chemiluminescence kit (Amersham, Little Chalfont, UK).

The tissues were fixed in formalin, embedded with paraffin and sliced into 4-µm sections for immunohistochemical staining. Three sections matched different levels of the lungs were collected and stained with hematoxylin and eosin (H&E). Then the lung metastasis number of cancers were counted.

miR-381 lentiviral expression or control vector were obtained from the company of GeneCopoeia. LRH-1 overexpression plasmid or control plasmid were purchased from Ruibo (Guangzhou, China). The tranfection of these vectors into NSCLC cells were performed in 6-well plates with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) based on the instruction from manufacturer.

The migration and invasion ability of NSCLC cells were investigated by Transwell assays without or with matrigels. Generally, NSCLC suspended in basal DMEM were seeded into the upper chamber and the lower chambers contained 600 µl DMEM with 20% FBS. After 24–48 h, NSCLC cells migrated or invaded through the membranes and stayed on the lower surface were stained with crystal violet. Cell number for the migrated or invaded cells was counted under a microscope.

Wild-type LRH-1 3′-UTR sequence and the mutated LRH-1 3′-UTR sequence were constructed into the pGL3 control vector (Promega, Madison, WI, USA) to get wt LRH-1-3′-UTR vector and mt LRH-1-3′-UTR vector, respectively. For luciferase reporter assay, NSCLC cells were co-transfected with the wild-type construct or mutant construct, and, miR-381 lentiviral expression or control vector. After 48 h of transfection, cells were harvested and lysed. The Dual-Luciferase Reporter Assay system (Promega, Shanghai, China) were used to the firefly and Renilla luciferase activities.

Data are presented as the mean ± SD and performed in at least three independent replicates. SPSS software, 16.0 (SPSS, Inc, Chicago, IL, USA) and Graphpad Prism 6.0 (CA, USA) were used for a two-tailed Student's t-test, Pearson's correlation analysis, Kaplan-Meier method and the log-rank test to evaluate the statistical significance. Differences were defined as P<0.05.

To determine the expression level of miR-381 in NSCLC, we performed qRT-PCR to detect miR-381 expression in 124 paired NSCLC tissues and adjacent non-tumor tissues. The results showed that miR-381 was significantly decreased in NSCLC tissues compared to adjacent non-tumor tissues (P<0.05, Fig. 1A). Besides, miR-381 expression was significantly downregulated in all NSCLC cell lines (A549, H1299, 95D, SPC-A1 and H358) compared with normal lung epithelial cell line BEAS-2B (P<0.05, Fig. 1B). Taken together, these data suggest that reduced miR-381 may be associated with NSCLC carcinogenesis and development.

To explore the function of miR-381 in NSCLC, we classified two different miR-381 groups according to median expression level to correlate miR-381 expression with clinicopathological characteristics in NSCLC patients. As shown in Table I, the downregulated miR-381 was significantly associated with advanced TNM stage (P=0.006) and lymph node metastasis (p=0.001). These data indicated that the reduced miR-381 was correlated with poor prognostic characteristics of NSCLC. Moreover, we determined the effects of miR-381 by Kaplan-Meier analysis curve and it showed that the higher miR-381 expression had significantly better overall survival (P=0.0001, Fig. 2). Furthermore, miR-381 was an independent factor for predicting 5-year overall survival in NSCLC patients (P=0.001, P=0.002, respectively, Table II). These data confirmed the potential value of miR-381 as a prognostic biomarker for NSCLC patients.

To clarify the biological function of miR-381 in NSCLC, we transfected miR-381 lentiviral expression or control vector into A549 (P<0.05, Fig. 3A) and H1299 cells (P<0.05, Fig. 3B) to establish two stable overexpression-miR-381 cell lines. As a result, overexpression of miR-381 inhibited cell migration (P<0.05, Fig. 3C) and invasion (P<0.05, Fig. 3D) as measured by Transwell assays. Besides, to prove the role of miR-381 in lung metastasis, we constructed lung metastasis mouse model by injecting LV-miR-381 or LV-control A549 or H1299 cells into tail vein of nude mice to induce pulmonary metastasis. Subsequently, overexpressing miR-381 significantly reduced the lung metastasis of A549 or H1299 cells (P<0.05, Fig. 3E). These data suggest that miR-381 played a key role in migration and invasion of NSCLC cells in vitro and in vivo.

To identify the target and possible mechanism of miR-381, we searched the candidate target genes of miR-381 by analyzing different bioinformatics algorithm (TargetScan and miRanda) and found that the 3′-UTR of LRH-1 contains a putative binding site for miR-381 (Fig. 4A). To validate that LRH-1 was a direct target of miR-381, a luciferase assay was conducted to determine whether miR-381 directly targets LRH-1 3′-UTR. The results showed that overexpression of miR-381 decreased the luciferase activity of the wild-type (wt) LRH-1 3′-UTR, but not mutant (mt) LRH-1 3′-UTR (P<0.05, Fig. 4B). Furthermore, miR-381 overexpression markedly suppressed LRH-1 mRNA (P<0.05, Fig. 4C) and protein (P<0.05, Fig. 4D) expression in A549 and H1299 cells. These results suggest that miR-381 could directly target LRH-1 in NSCLC cells through binding its 3′-UTR.

To further confirm the above data which suggest LRH-1 was a direct target of miR-381, we performed qRT-PCR to explore the LRH-1 expression in NSCLC tissues. Our data revealed that LRH-1 mRNA and protein levels were significantly lower in high miR-381 group than that in low miR-381 group in NSCLC (P<0.05, Fig. 5A and B). Moreover, the results showed that the mRNA level of LRH-1 in the NSCLC tissues was inversely correlated with miR-381 expression (R2=0.6248, P<0.0001, Fig. 5C). Taken together, these data indicated that LRH-1 was a direct downstream target of miR-381 in NSCLC tissues.

To further confirm LRH-1 was a functional target and was involved in miR-381-mediated biological effects, we restored LRH-1 expression in A549-miR-381 and H1299-miR-381 cells by transfecting LRH-1 expression plasmid (P<0.05, Fig. 6A). Furthermore, Transwell assay showed that LRH-1 overexpression promoted cell migration (P<0.05, Fig. 6B) and invasion (P<0.05, Fig. 6C), which abolished the suppressive effects of miR-381 on A549 and H1299 cells. These data demonstrated that LRH-1 mediates the functional effects of miR-381 on migration and invasion in NSCLC cells.