Specimens of lung adenocarcinoma and paracarcinoma normal tissues were collected from 14 patients (7 males and 7 females, aged 40–59 years, 4 patients in stage IB and 10 patients in stage IIB) who were pathologically diagnosed with lung adenocarcinoma at YantaiShan Hospital (Yantai, China) between March 5 and August 31, 2015. All patients were diagnosed for the first time and had not received chemotherapy. Fresh tissues from the patients were prepared for RNA analysis immediately following surgery. All experiments were performed in accordance with the relevant guidelines and approved by the Medical Ethics Committee of Binzhou Medical University (Yantai, China). Prior to study inclusion, written informed consent was obtained from all patients.

Specimens of lung adenocarcinoma and paracarcinoma tissues from 14 patients, and xenograft tumors from nude mice, were collected and ground into powder in liquid nitrogen; the cultured cells did not need to be homogenized. miRNAs from lung adenocarcinoma cells or tissues were isolated using a miRNA kit (Takara Bio, Inc., Otsu, Japan). After measuring the concentration of miRNAs, poly(A) was added using poly(A) polymerase (Ambion; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Then, PrimeScript™ RT reagent kit with gDNA Eraser (Takara Bio, Inc.) was used to perform the RT reaction with primer [5′-AACATGTACAGTCCATGGATGd(T)30N(A, G, C or T)-3′]. The SYBR Premix Ex Taq kit (Takara Bio, Inc.) was used to perform qPCR with the 7500 Fast Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). The qPCR conditions were as follows: initial denaturation at 95°C for 30 sec; 40 cycles of 95°C for 10 sec, 60°C annealing for 20 sec and extension at 72°C for 20 sec. Then, fluorescence was detected at 585 nm. The primers used to amplify miR-33a-5p were 5′-GTGCATTGTAGTTGCATT-3′ (forward) and 5′-AACATGTACAGTCCATGGATG-3′ (reverse). The primers for 5S rRNA were 5′-GCCATACCACCCTGAACG-3′ (forward) and 5′-AACATGTACAGTCCATGGATG-3′ (reverse). The human 5S rRNA gene served as the control. The results were calculated using the 2^−ΔΔCq value (20).

miR-33a-5p mimics and negative control (nc) oligonucleotides were chemically synthesized. The sequences of the miR-33a-5p mimics were as follows: GUGCAUUGUAGUUGCAUUGCA (sense) and UGCAAUGCAACUACAAUGCACUU (antisense). The sequences of the nc oligonucleotides were as follows: CAGUACUUUUGUGUAGUACAA (sense) and GUACUACACAAAAGUACUGUU (antisense). These sequences were inserted into the pGCMV/EGFP/miR/blasticidin vector during the construction of the miRNA overexpression vector. This part of the experiment was performed by GenePharma Biotech Co., Ltd. (Shanghai, China).

Lung adenocarcinoma (LTEP-a-2 and A549) cells and human bronchial epithelial (HBE) cells, which are all adherent cells, were maintained in 1640 medium supplemented with 10% fetal bovine serum (FBS) (both from Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C with 5% CO₂ under saturation humidity. The synthesized miRNA mimics were transfected when cells had reached 50–60% confluence. A total of 1 μg miRNA mimics was mixed with 3 μl Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), in accordance with the manufacturer's instructions. Cells were treated with 50 μM miRNA mimics. Blasticidin (Solarbio Science and Technology Co., Ltd., Beijing, China) was used to select antibiotic-resistant cells and to detect cells that expressed miR-33a-5p stably. Cells with stable miR-33a-5p expression were used for the xenograft experiments in mice. In addition, 0.75 μM celastrol (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) and 25 μM miRNA was used to detect the effects of combined treatment. Celastrol was added when changing the calf serum medium 6 h after transfection.

Logarithmic-phase cells (1×10⁴) in each well of 96-well plates were treated with miRNA or celastrol for 48 h. At 4 h before the end of incubation, 10 μl MTT (5 mg/ml; Sigma-Aldrich; Merck KGaA) was added to each well. Then, 100 μl DMSO (Sigma-Aldrich; Merck KGaA) was added, and the plate was shaken until MTT was dissolved. The optical density (OD) was measured at 490 nm using an auto-microplate reader (Thermo Fisher Scientific, Inc.) to compare the proliferation of each group of cells.

Cells were treated with miRNA or celastrol for 48 h, and the culture medium was discarded. The cells were digested from the bottom of the culture flask with 0.25% trypsin enzyme without EDTA, then centrifuged at 100 × g for 5 min. PBS was used to wash the cells. The cellular apoptosis ratio was detected using Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) staining (BD Pharmingen; BD Biosciences, Franklin Lakes, NJ, USA). Annexin V-FITC (5 μl) was added to the collected cells. After complete mixing, 5 μl PI was added. Finally, 1×10⁴ cells were analyzed using a flow cytometer (Beckman Coulter, Inc., Brea, CA, USA). The data analysis was performed using the CytExpert 1.2.11.0 software (Beckman Coulter, Inc.).

Cells were treated with miRNAs or celastrol for 24 h and collected. Then, the cells were seeded into the upper chamber (10⁵ cells/well in 400 μl 1640 medium, FBS-free) of Transwells (Corning Inc., Corning, NY, USA). The lower chamber was filled with 600 μl 1640 medium supplemented with 20% calf serum (Gibco; Thermo Fisher Scientific, Inc.). After 24 h, the liquid in the upper chamber was removed with wet swabs, and the upper surface was carefully washed with methanol to fix the cells. Then, the cells that had traversed the membrane were stained with 0.1% crystal violet and counted under an inverted light microscope (Leica Microsystems GmbH, Wetzlar, Germany). The migration capability of the cells in vitro was assessed according to the number of transmembrane cells. Average trans-membrane cells number were determined in five ×200 fields.

Cells were lysed with lysis buffer (21), and all proteins were collected. The xenograft tumors were collected, ground into powder in liquid nitrogen, then lysed with lysis buffer to collect all proteins. Then, 40 μg protein was loaded into individual lanes and separated via SDS-PAGE. Subsequently, the proteins were transferred onto polyvinylidene fluoride membranes (Sigma-Aldrich; Merck KGaA), which were blocked with 7% nonfat milk in TBST for 2 h. After washing with TBST, the membranes were incubated with rabbit anti-human mTOR (1:800; cat. no. BS3611; Bioworld Technology, Nanjing, China), rabbit anti-human phosphorylated (p)-p70 ribosomal protein S6 kinase (p70S6K; 1:500; cat. no. BS4439), rabbit anti-human p-eukaryotic translation initiation factor 4E binding protein 1 (4EBP1; 1:500; cat. no. BS4746), and rabbit anti-human GAPDH (1:6,000; cat. no. AP0063) (all from Bioworld Technology) in TBST at 4°C overnight. Horseradish peroxidase-labeled goat anti-rabbit Immunoglobulin G (1:6,000; cat. no. ZB-2301; Beijing ZhongShan Golden Bridge Technology Co., Ltd., Beijing, China) was added, and the samples were incubated for 2 h at room temperature. Finally, images of the membranes were captured using a chemiluminescent imager (Tanon Science and Technology Co. Ltd., Shanghai, China). The densities of the bands were analyzed using Gel Image System 4.2 software (Tanon Science and Technology Co. Ltd.).

mTOR-3′-UTR double-stranded DNA (203 bp) containing an incomplete matched area of miR-33a-5p was synthesized and inserted into the dual-luciferase reporter pmirGLO vector (Promega Corp., Madison, WI, USA) by SacI/XhoI dual-enzyme digestion (Takara Bio, Inc.). Thus, GP-miRGLO-mTOR-WT was constructed. In addition, the nucleotide sequence was altered to construct the GP-miRGLO-mTOR-MUT vector as the control. A549 and LTEP-a-2 cells were transfected with miR-33a-5p and the dual-luciferase reporter pmirGLO vector, which simultaneously expressed the Firefly and Renilla luciferases. Cells were collected after 48 h incubation, and 100 μl 1X passive lysis buffer was added to each well. Then, 20 μl sample and 100 μl Luciferase Assay Reagent II were added to each well of 96-well white flat bottom plates. The activity of Firefly lucif-erase in each well was detected using a luminescent detection system (Tecan Group, Ltd., Mannedorf, Switzerland) read as M1. Then, 100 μl 1X Stop & Glo reagent was added to each well. The activity of Renilla luciferase was detected using the luminescent detection system, read as M2. The ratio of M1/M2 was calculated, and the relative luciferase activity of each group was evaluated.

Cells were transfected with miR-33a-5p overexpression vector and selected with blasticidin. A total of 16 BALB/c-nu/nu 5-6 week-old male mice with an average weight of 18-20 g (Charles River, Beijing, China) were randomly divided into 4 groups of 4 mice in each. They were kept in a laminar airflow cabinet under specific pathogen-free conditions with a controlled temperature (23±2°C), humidity (40–70%) with free access to food and water. The cells were transplanted subcutaneously into the right or left flanks of these mice. Tumors appeared after ~3 days. The tumor volume was measured daily with calipers, with the following formula: Tumor volume (mm³) = A x B²/2, where 'A' and 'B' are the maximum and minimum tumor diameters, respectively. When the tumor volume of the control group increased to ~150 mm³, half of the mice from the control and the miRNA-overexpressing groups were randomly selected. Celastrol was injected into these selected mice at 2 mg/kg/day, 5 days/week. The other half of the groups was intraperitoneally injected with saline as a control. Following drug treatments for 6 weeks, all of the mice were sacrificed for tumor isolation by cervical vertebra dislocation. The tumors were then collected and weighed. Subsequently, all of the tumors were used for RNA extraction and total protein detection. All animal experiments were approved by the Committee on the Ethics of Animal Experiments of Binzhou Medical University and conducted based on the National Institutes of Health (Bethesda, MD, USA) Guide for the Care and Use of Laboratory Animals.

SPSS v22.0 software (IBM Corp., Armonk, NY, USA) was utilized for statistical analysis. Student's t-test and one-way analysis of variance followed by a Tukey's test were used to compare variables among groups. P<0.05 was considered to indicate a statistically significant difference.

The expression of miR-33a-5p was measured in lung tumors and the corresponding adjacent tissues to examine a potential role of miR-33a-5p on the development of lung adenocarcinoma. The results demonstrated that the expression levels of miR-33a-5p were significantly lower in lung cancer tissues compared with the corresponding adjacent tissues (Fig. 1A). When examining cell lines, the expression levels of miR-33a-5p were lower in the lung adenocarcinoma cell lines LTEP-a-2 and A549 compared with the normal HBE cells (Fig. 1B). Thus, downregulation of miR-33a-5p expression is likely to be involved in the development of lung adenocarcinoma.

Considering the decrease in miR-33a-5p expression in lung adenocarcinoma tissues and cells, the effect of miR-33a-5p on the proliferation of lung adenocarcinoma cells was examined in vitro. At 48 h following transfection of miR-33a-5p mimics into A549 and LTEP-a-2 cells, a lower number of cells was observed in the miR-33a-5p overexpression group compared with the scrambled control group, in both the cell lines tested (Fig. 2A and B). In addition, MTT assay was used to measure cell viability. The OD value of the overexpression group was decreased compared with the scrambled control group (Fig. 2C and D), which indicated that cell proliferation in the overexpression group was inhibited. The results of the Transwell cell migration assays demonstrated that the migration ability of A549 and LTEP-a-2 cells was significantly inhibited following miR-33a-5p overexpression (Fig. 2E and F). Furthermore, miR-33a-5p overexpression enhanced apoptosis; the % of apoptotic cells was over four times higher in the miR-33a-5p overexpression group compared with the scrambled control group for both the A549 (Fig. 2G) and LTEP-a-2 (Fig. 2H) cells.

A miRNA could inhibit the proliferation of lung adenocarcinoma cells by modulating the expression of its target genes. Therefore, predicted target genes were examined for miR-33a-5p by bioinformatics analysis (www.microrna.org/microrna/getMirnaForm.do and www.targetscan.org/index.html) and matching sites were identified on the 3′-UTR of the mTOR gene (Fig. 3A). According to the mechanism of miRNA function, miR-33a-5p likely affects the expression of mTOR by binding to the mTOR-3′-UTR. Wild-type (WT) mTOR-3′-UTR was cloned downstream of the Firefly luciferase gene to construct the GP-miRGLO-mTOR-WT dual-luciferase vector. In addition, the predicted binding sites (852-863 bp) were replaced with the complementary sequences to construct the mutant (MUT) GP-miRGLO-mTOR-MUT vector (Fig. 3B). The cells were transfected with the GP-miRGLO-mTOR-WT or GP-miRGLO-mTOR-MUT vector, together with miR-33a-5p mimics or scrambled control. Luciferase activity was detected after 48 h incubation. The results demonstrated that the relative luciferase activity significantly decreased in the cells transfected with the GP-miRGLO-mTOR-WT vector and miR-33a-5p mimics (Fig. 3C and D). No notable difference was observed between cells transfected with GP-miRGLO-mTOR-MUT vector and miR-33a-5p mimics (Fig. 3C and D).

The effect of miR-33a-5p transfection on mTOR expression in lung adenocarcinoma cells was determined. RT-qPCR analysis confirmed that the expression levels of miR-33a-5p significantly increased in the A549 and LTEP-a-2 cells following mimics transfection (Fig. 4A and B). The results from the western blot analysis of the transfected cells indicated that the protein expression levels of mTOR were significantly inhibited following miR-33a-5p overexpression (Fig. 4C–F). Furthermore, the protein expression levels for the downstream effectors of mTOR, p-p70S6K and p-4EBP1, were significantly decreased following miR-33a-5p overexpression (Fig. 4C–F). These results indicated that miR-33a-5p inhibited the proliferation of lung adenocarcinoma cells via the mTOR signaling pathway.

Celastrol treatment was combined with miR-33a-5p overexpression and microscopic observation of the cells revealed that the number of visible living cells was significantly reduced in the combination treatment group compared with the negative control group, and the single-use 0.75 μM celastrol or 25 nM miR-33a-5p groups (Fig. 5A and B). In the MTT assay, the OD of the combination treatment group was lower compared with the other groups, which indicated a reduction in cell viability (Fig. 5C and D). Therefore, the combination treatment inhibited cell proliferation more efficiently than either treatment alone. The % of apoptotic cells in the A549 and LTEP-a-2 cells was ~50% in the combination treatment group, but <20% in the single treatment groups (Fig. 5E and F).

The expression levels of miR-33a-5p were examined by RT-qPCR, in order to explore the mechanism by which miR-33a-5p enhances the sensitivity of lung adenocarcinoma cells to celastrol. Although miR-33a-5p expression was increased in the celastrol alone group, it was considerably higher in the celastrol and miR-33a-5p combination treatment group (Fig. 6A and B). Furthermore, results from western blot analysis demonstrated that the expression of mTOR was markedly decreased in the combination treatment group as the expression of miR-33a-5p increased (Fig. 6D–F). Similarly, p-p70S6K and p-4EBP1 expression was also further decreased in the combination treatment group (Fig. 6D–F). These results suggested that mTOR was involved in the mechanism by which miR-33a-5p enhanced the sensitivity of lung adenocarcinoma cells to celastrol. The results were similar in A549 (Fig. 6C and E) and LTEP-a-2 (Fig. 6D and F) cells.

Parental untreated LTEP-a-2 cells (negative) or LTEP-a-2 cells stably transfected with miR-33a-5p overexpression vector (miR-33a-5p) were injected into the flanks of nude mice. Mice were then administered with celastrol or saline treatment to verify that the combination of celastrol and miR-33a-5p could inhibit LTEP-a-2 cell growth in vivo. The growth curve of the xenograft tumors revealed that the tumor volume increased with prolonged incubation time in the negative control group. The tumor growth rate was considerably lower in the celastrol treatment alone group or the miR-33a-5p overexpression alone group compared with the negative control group. In the combination treatment group, the tumor growth was almost completely arrested after 2 weeks (Fig. 7A). The xenograft tumors were extracted at the end of the experiment and the volumes and weight measurements of the xenograft tumors varied among the different groups (Fig. 7B and C). The xenograft tumors in the combination treatment group were the smallest (Fig. 7B and C). RT-qPCR analysis of the xenograft tumors indicated that the cells stably transfected with the miR-33a-5p overexpression vector maintained a higher expression level of miR-33a-5p compared with the parental cells (Fig. 7D). Celastrol treatment further enhanced the expression of miR-33a-5p (Fig. 7D), which was consistent with the in vitro experiments. Finally, similarly to the in vitro experiments, higher expression levels of miR-33a-5p resulting from the combination treatment significantly inhibited the expression of mTOR and its downstream effectors in the xenograft tumors (Fig. 7E and F).