Lung cancer specimens and adjacent normal tissues (n=70) were obtained from patients suffering from NSCLC who had not received any pre-operative radiotherapy and chemotherapy at Wujin People’s Hospital of Changzhou (Changzhou, China) from January, 2010 to May, 2013. The patients’ sex distribution (male, n=39; female, n=31) and age distribution in years (≥60, n=38; <60, n=32; age range, 27 to 79 years; mean age, 58.88±12.65 years) were recorded. The diagnosis of lung cancer was performed by our surgeons and pathologists. The collected tissues were stored at −80°C in liquid nitrogen until use. Tissues and information of patients were obtained with written informed consent. The procedures used in this study were approved by the Human Ethics Committee at the Wujin People’s Hospital of Changzhou. All the experiments were conducted according to the approved guidelines.

The human NSCLC A549 (no. CCL-185), H1299 (no. CRL-5803), ANIP-973 (no. CS-0168) and GLC-82 (no. ZY-H065) cell lines (American Type Culture Collection, Manassas, VA, USA) were cultured in DMEM (Gibco/Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS) and antibiotics at 37°C in a humidified atmosphere containing 5% CO₂.

A total of 10 male NOD/SCID mice (6-8 weeks old) were purchased from HFK Bio-Technology Co. (Beijing, China). Mice were housed under specific pathogen-free (SPF) conditions and were quarantined for 1 week prior to treatment. The animals were housed in a controlled environment with a temperature of 20-22°C, relative humidity of 50-60% and 12 h light/dark cycles, and were provided with access to food and water ad libitum. All animal procedures were conducted in accordance with the protocol approved by the Institutional Animal Care and Treatment Committee of Wujin People’s Hospital of Changzhou. All animals were treated humanely throughout the experimental period.

Total RNA was extracted from the collected tissues or cells using TRIzol reagent (Invitrogen/Thermo Fisher Scientific, Inc.) according to the manufacturer’s instructions. RNA was converted into cDNA using a Reverse Transcription kit (Thermo Fisher Scientific, Inc.). PCR was then performed using Taq polymerase (Takara Bio, Inc., Otsu, Japan). Stem-loop primers (Guangzhou RiboBio Co., Ltd., Guangzhou, China) were used to detect miRNAs. U6 small nucleolar RNA and GAPDH were used for normalization. The relative expression levels were calculated using the 2^−ΔΔCt method (32) (CFX manager software 3.1; Bio-Rad Laboratories, Inc., Hercules, CA, USA). The primers used were as follows: hsa-miR-98-5p forward, 5′-ACACTCCCUAUACAACUUAC-3′ and reverse, 5′-GGGAAAGUAGUGAGGCCTCAGA-3′; TGFBR1 forward, 5′-TCGTCTGCATCTCACTCAT-3′ and reverse, 5′-GATAAATCTCTGCCTCACG-3′; and GAPDH forward, 5′-TCTCTGCTCCTCCTGTTC-3′ and reverse, 5′-GGTTGAGCACAGGGTACTTTATTGA-3′. Quantitative PCR was carried out using SYBR Premix Ex Taq II (Tli RNaseH Plus; 2X; 6 µl), 0.5 µl PCR forward primer (10 µM), 0.5 µl PCR reverse primer (10 µM), 1 µl cDNA solution, 3 µl RNase-free dH₂O, total was 10 µl. The cycling conditions for PCR were as follows: Stage 1: 95°C, 30 sec; repeat: 1; stage 2: 95°C, 5 sec, 60°C, 34 sec; repeat: 40; stage 3: dissociation.

Hsa-miR-98-5p mimics, hsa-miR-98-5p inhibitors and the corresponding negative controls were purchased from Obio Technology (Shanghai, China) and transfected into the A549, H1299, ANIP-973 and GLC-82 cells using Lipofectamine^® 2000 (Invitrogen/Thermo Fisher Scientific, Inc.) according to the manufacturer’s instructions. At 48 h following transfection, the cells were collected for use in further experiments.

Cell viability was assessed using an MTT assay. The cells were seeded into 96-well plates at a density of 2,000 cells/well. After 12 h, the cells were transfected with miR-98-5p mimics or miR-NC. At 24, 48, 72 or 96 h following transfection, 20 µl MTT solution (5 mg/ml; Amresco, Inc., Framingham, MA, USA) was added to each well followed by incubation at 37°C for 4 h. The cells were subsequently treated with 200 µl DMSO to dissolve the precipitated crystals. Finally, the absorbance of each well was measured using a microplate reader (EL10A; Biobase, Jinan, China) at 570 nm. All experiments were performed 3 times. Data are presented as the means ± standard deviation (SD).

The cells were seeded in 6-well plates (500 cells/well) and incubated at 37°C for 12 h. Following transfection with miR-98-5p mimics or miR-NC, the cells were further incubated at 37°C for 6-8 days. The cells were fixed with methanol for 10 min and stained with 0.1% crystal violet at room temperature for 30 min. Images were subsequently captured using a microscope (Olympus, Tokyo, Japan).

The cell migratory and invasive abilities were investigated using a 24-well plate with 8-µm pore size Boyden chamber inserts. For the migration assays, 4×10⁴ cells were seeded into the upper chamber without a coated membrane. For the invasion assays, 1×10⁵ cells were seeded into the upper chamber with a Matrigel-coated membrane. The cells in the upper chamber were transfected with miR-98-5p mimics or inhibitors and suspended in 200 µl serum-free DMEM. The control groups were transfected with corresponding negative controls. A total of 800 µl DMEM supplemented with 10% FBS was added to the lower chamber. The chambers were incubated for 24 h at 37°C and the non-migrated and non-invasive cells remaining in the upper chamber were removed using a cotton swab. Migrated or invaded cells in the lower chamber were fixed with methanol and stained with 0.1% crystal violet at room temperature for 20 min. Finally, the cells were imaged by the microscope (Olympus) and counted using high-power magnifications.

The cells were seeded in 6-well plates (5×10⁵ cells/well) and incubated at 37°C for 12 h. When the cells reached 90% confluence, a scratch wound was made using a 10-µl pipette tip. The cells were subsequently washed with PBS to remove cell debris and transfected with miR-98-5p mimics or inhibitors. The control groups were transfected with corresponding negative controls. The cells were incubated with DMEM supplemented with 1% FBS for 24 h, following which images were captured under a microscope (Olympus).

Western blot assays were performed to measure the expression of target proteins. Total cellular proteins were extracted from the cultured cells using lysis buffer [20 mM Tris (pH7.4), 150 mM NaCl, 5 mM EDTA, 50 mM NaF and 0.1% NP-40]. Proteins were separated by SDS-PAGE (10%, w/v) and transferred onto PVDF membranes, which were then blocked using 5% skim milk for 2 h at room temperature. The membranes were subsequently incubated with target primary antibodies against the following: MMP-9 [13667S; Cell Signaling Technology (CST), Danvers, MA, USA], vimentin (10R-1903; Fitzgerald Industries International, Acton, MA, USA), α-smooth muscle actin (α-SMA; 19245; CST), TGFBR1 (70R-36484) and β-actin (10R-10263) (both from Fitzgerald Industries International) at 4°C overnight. The dilution used for MMP-9, vimentin, α-SMA and TGFBR1 was 1:100, and the dilution used for β-actin was 1:1,000. Following incubation with horseradish peroxidase-labeled secondary antibodies (A0216; Byotime, Shanghai, China) which were diluted at 1:1,000 at 37°C for 2 h, the protein bands were visualized using a chemiluminescence (ECL) detection system and images were captured.

The 3′-UTR of wild-type (WT) and mutant TGFBR1 were amplified from human genomic DNA and individually inserted into pmiR-RB-REPORT™ luciferase vectors (Obio, Shanghai, China). The A549 cells were co-transfected with 200 ng of mutant or WT pmiR-RB-REPORT™ plasmid and 100 ng of miR-98-5p mimics or miR-NC Following incubation for 36 h, and luciferase activity was measured with a Dual-Luciferase Reporter Assay System (Promega Corporation, Madison, WI, USA) according to the manufacturer’s instructions.

A549 cells suspensions (1×10⁷ cells; 200 µl) were subcutaneously injected into the right flanks of NOD/SCID mice to establish a xenograft NSCLC model. After 6 days, tumor volume reached ~200 mm³, and the mice were randomly divided into 2 groups as follows: The control (n=5) and miR-98-5p mimic (n=5) group. A total of 5 µg/mouse plasmid was administered every 3 days. The treatment was terminated when mice in the control group became moribund (at day 21). Mice were sacrificed, and tumors were harvested, weighed and imaged. Finally, tumors were collected for further hematoxylin and eosin (H&E) and immunohistochemical evaluation. Tissues were collected, fixed in 4% paraformaldehyde, embedded in paraffin and sectioned (5 µm thickness). After dewaxing and rehydration, the sections were stained with H&E at room temperature for 10 sec and imaged under a light microscope (Olympus). For immunohistochemistry, the sections were respectively incubated with primary rat anti-mouse antibody Ki-67 (9449; CST; dilution: 1:400), MMP-9 (13667S; CST; dilution: 1:325) and TGFBR1 (70R-36484; Fitzgerald Industries International; dilution: 1:200), and then treated with secondary antibody biotinylated goat anti-rat immunoglobulin (Abcam, Cambridge, MA, USA). The sections were incubated with streptavidin-peroxidase and DAB solution (Solarbio, Beijing, China) at 37°C for 20 min to visualize the biotinylated goat anti-rat immunoglobulin. The sections were then treated with hematoxylin at room temperature for 10 sec to stain the nuclei of the tumor cells. Finally, the sections were imaged and examined under a light microscope (Olympus).

A total of 1×10⁶ of A549 stable luciferase reporter cells suspended in 100 µl PBS were injected into NOD/SCID mice via the tail vein to establish a model of pulmonary metastasis. At 3 days after inoculation, 10 mice were randomly divided into the control and miR-98-5p mimics groups (n=5) and treated with 5 µg miR-98-5p mimic plasmids or control plasmids, respectively, every 3 days. At 5, 10 and 15 days following inoculation, mice were administered with D-luciferin (3 mg/mouse) for analysis. Bioluminescence was then detected from lung metastatic tumors was detected using an IVIS Lumina In Vivo Imaging System (Perkin-Elmer, Waltham, MA, USA). Finally, the lungs were harvested from each mouse to count the metastatic nodes and fixed with 4% paraformaldehyde for further H&E evaluation.

The results are presented as the means ± SD. Statistical analysis was performed using SPSS 17.0 software (SPSS Inc., Chicago, IL, USA). The significance among multiple groups was evaluated using one-way ANOVA followed by Newman-Keuls (SNK) t-test. Significant differences between 2 groups (parametric) were analyzed using a Student’s t-test, and significant differences between 2 groups (non-parametric) were analyzed by the Mann-Whitney U test. A value of P<0.05 was considered to indicate a statistically significant difference.

Lung cancer progression is a complex process associated with the activation of oncogenes and the silencing of tumor suppressor genes (32-34). It has been reported that miRNA dysregulation is significantly associated with tumorigenesis and cancer progression (35). In this study, to confirm the biological function of miR-98-5p in NSCLC, we first detected its expression in 70 human primary lung tumor tissues and adjacent normal lung tissues by RT-PCR. The results revealed that miR-98-5p expression was significantly downregulated in the tumor tissues (60.3%) compared with in the controls (Fig. 1A and B; P<0.01). In addition, the expression of miR-98-5p in patients with advanced NSCLC (stage III; n=41) was significantly decreased compared with that in patients with the early stage NSCLC (stage I; n=29; P<0.05; Fig. 1C). Furthermore, compared with non-metastatic NSCLC (n=25), miR-98-5p expression was distinctly downregulated in metastatic NSCLC (n=45; P<0.05, Fig. 1D). These results suggest that miR-98-5p is downregulated in NSCLC and is associated with NSCLC progression.

In order to investigate the biological role of miR-98-5p in NSCLC, we measured the expression of miR-98-5p in 5 different NSCLC cell lines. It was observed that miR-98-5p expression was downregulated in the A549 and H1299 cells, but was increased in the ANIP-973 and GLC-82 cells (Fig. 2A). We thus transfected the A549 and H1299 cells with plasmid miR-98-5p mimics to restore miR-98-5p expression, and the results revealed that miR-98-5p expression was significantly increased following transfection (P<0.001; Fig. 2B and C). Following transfection with miR-98-5p mimics, the viability (Fig. 2D and E) and colony formation abilities (Fig. 2F and G) of the A549 and H1299 cells were significantly suppressed, indicating an anti-proliferative effect of miR-98-5p mimics. A number of studies have demonstrated that miRNA dysregulation is associated with tumor migration and invasion (33,34), and we thus used miR-98-5p mimics to restore the expression level of miR-98-5p in the A549 and H1299 cells. The results of a wound healing assay revealed that the migration of the A549 and H1299 cells was suppressed following transfection with miR-98-5p mimic (P<0.05; Fig. 3A). Furthermore, Transwell assays revealed that the migration and invasion of A549 and H1299 cells were distinctly suppressed by transfection with miR-98-5p mimics (P<0.001). The expression of proteins associated with tumor metastasis and EMT was measured using western blotting following treatment with miR-98-5p mimics. As shown in Fig. 3B and C, the expression of MMP-9, α-SMA and vimentin was significantly downregulated in the A549 and H1299 cells following transfection with miR-98-5p mimics. These results suggested that the restoration of miR-98-5p expression significantly suppressed the viability, proliferation, migration and invasion of both the A549 and H1299 cells, acting as a tumor suppressor in vitro.

To further confirm that miR-98-5p acts as a tumor suppressor in NSCLC, we utilized a plasmid miR-98-5p inhibitor. Compared with the miR-controls, transfection with miR-98-5p inhibitors significantly decreased the expression of miR-98-5p in both the ANIP-973 and GLC-82 cells (Fig. 4A; P<0.001). When miR-98-5p expression was downregulated, the viability (Fig. 4B and C) and colony formation ability (Fig. 4D and E) of the ANIP-973 and GLC-82 cells were increased. Furthermore, cell migration and invasion were investigated following the knockdown of miR-98-5p. It was demonstrated that the mobility of the ANIP-973 and GLC-82 cells was increased following miR-98-5p downregulation (Fig. 5A). In addition, the migration and invasion of the ANIP-973 and GLC-82 cells were distinctly enhanced following transfection with miR-98-5p inhibitors. When the ANIP-973 and GLC-82 cells were transfected with the miR-98-5p inhibitors, the expression levels of metastasis-associated proteins (MMP-9, α-SMA and vimentin) were upregulated, which was in agreement with the wound healing and Transwell assay results (Fig. 5B and C). Taken together, these results suggest that miR-98-5p downregulation enhances the viability, proliferation, migration and invasion of NSCLC cells; in other words, miR-98-5p inhibits NSCLC tumorigenesis in vitro.

In order to further investigate the mechanisms through which miR-98-5p affects NSCLC cell progression, we used the computational prediction program TargetScan (www.targetscan.org) to predict target genes. In total, 60 genes were simultaneously predicted by the TargetScan databases, and TGFBR1 was detected as a candidate gene related to NSCLC based on its associated Gene Ontology (GO) terms and also harbors miR-98-5p binding sites, suggesting that TGFBR1, which stimulates tumor cell proliferation, migration and invasion via the TGF-β/Smad signaling pathway (35-38), was identified as a potential target of miR-98-5p (Fig. 6A).

To confirm that TGFBR1 is a downstream target of miR-98-5p, we transfected the A549 cells with miR-98-5p mimics and found that TGFBR1 expression was significantly decreased (P<0.001; Fig. 6B and C). Conversely, TGFBR1 expression was upregulated in the GLC-82 cells following miR-98-5p knockdown (P<0.001; Fig. 6D and E). A dual-luciferase reporter assay was also performed to further verify that miR-98-5p regulates TGFBR1. The WT TGFBR1 3′-UTR (pmiR-RB-REPORT™-TGFBR1-3′-UTR-WT) and mutated TGFBR1-3′-UTR (pmiR-RB-REPORT™-TGFBR1-3′-UTR-MUT) were cloned and co-transfected with miR-98-5p or miR-NC into A549 cells. As shown in Fig. 6F, the luciferase activity of pmiR-RB-REPORT™-TGFBR1-3′-UTR-WT was distinctly inhibited by miR-98-5p mimics compared with miR-NC. However, no significant differences were observed in the luciferase activity of pmiR-RB-REPORT™-TGFBR1-3′-UTR-MUT in the presence of miR-98-5p mimics or miR-NC. These results suggest that miR-98-5p directly targets the 3′-UTR of TGFBR1.

Based on the anti-proliferative effects of miR-98-5p in vitro, these effects were further assessed using a xenograft NSCLC mouse model. Compared with the control group, mice in the miR98-5p mimics group exhibited a significantly slower tumor growth (P<0.01; Fig. 7A and B). Tumor weight was considerably reduced in the miR-98-5p mimic group (0.81±0.091 g) compared with the control group (1.13±0.082 g; P<0.01; Fig. 7C), indicating that miR-98-5p exerted a significant anti-tumor effect. Pathological analysis was performed to evaluate the anti-tumor effects of miR-98-5p (Fig. 7D). H&E staining of the tumor sections revealed preferable anti-proliferative effects in the miR-98-5p mimics group. The number of Ki-67-positive cells in the tumor sections was considerably lower in the miR-98-5p mimic group. Furthermore, TGFBR1 expression was upregulated in the control group compared with the miR-98-5p mimic group. The expression of MMP-9 was decreased in the miR-98-5p mimics group, which is in agreement with the results of immunohistochemical TGFBR1 staining. Taken together, these results suggest that miR-98-5p exerts significant anti-tumor effects in vivo.

A pulmonary metastasis model of A549-luciferase cells in NOD/SCID mice was used to evaluate the anti-metastatic effect of miR-98-5p mimics. Bioluminescence at the lung site, which indicates metastatic tumor cells, increased rapidly in the control group following implantation. However, bioluminescence remained largely unchanged in the miR-98-5p mimic treatment group. At 15 days after tumor cell implantation, bioluminescence was significantly lower in the miR-98-5p mimic group (37475±16967.87 p/sec/cm²/sr) compared with the control group (160432±16443.39 p/sec/cm²/sr; P<0.001). The harvested lung tissues were subjected to H&E staining and metastatic nodes were counted. As shown in Fig. 8D, the number of metastatic nodes in the miR-98-5p group was decreased compared with the control group. Furthermore, H&E staining of the lung tissue sections revealed fewer tumor nodes in the miR-98-5p group compared with the control group (Fig. 8C). These results suggest that miR-98-5p is able to effectively inhibit tumor metastasis in vivo.