Overexpression of miR-758 inhibited proliferation, migration, invasion, and promoted apoptosis of non-small cell lung cancer cells by negatively regulating HMGB

Non-small cell lung cancer (NSCLC) is one of the most fatal types of cancer with significant mortality and morbidity worldwide. MicroRNAs (miRs) have been confirmed to have positive functions in NSCLC. In the present study, we try to explore the role of miR-758 in proliferation, migration, invasion, and apoptosis of NSCLC cells by regulating high-mobility group box (HMGB) 3 (HMGB3.) NSCLC and adjacent tissues were collected. Reverse transcription quantitative PCR (RT-qPCR) was employed to detect expression of miR-758 and HMGB3 in NSCLC and adjacent tissues, in BEAS-2B cells and NSCLC cell lines. The targetted relationship between miR-758 and HMGB3 was identified by dual luciferase reporter gene assay. The effects of miR-758 on proliferation, migration, invasion, cell cycle, and apoptosis of A549 cells. MiR-758 expression was lower in NSCLC tissues, which was opposite to HMGB3 expression. The results also demonstrated that miR-758 can target HMGB3. The cells transfected with miR-758 mimic had decreased HMGB3 expression, proliferation, migration, and invasion, with more arrested cells in G1 phase and increased apoptosis. Our results supported that the overexpression of miR-758 inhibits proliferation, migration, and invasion, and promotes apoptosis of NSCLC cells by negative regulating HMGB2. The present study may provide a novel target for NSCLC treatment.


Introduction
Non-small cell lung cancer (NSCLC) is a prevalent form of lung cancer in the elderly, and its incidence is predicted to substantially increase [1]. As a form of a preventable cancer caused by tobacco use, NSCLC perhaps accounts for 85% of all lung cancer cases [2,3]. Recently, NSCLC has been reported to be related to molecular features which have critical implications in tumorigenesis as well as targetted treatments [4]. Individually customized chemotherapy is likely to improve the outcomes [5]. Before 2009, NSCLC was treated by cytotoxic chemotherapy which provided a 20-35% response rate and a 10-12 months median survival time [6]. The NSCLC patients treated with epidermal growth factor receptor inhibitors have an increased median survival time up to 18 months [7]. However, advanced NSCLC patients with poor progression have less response to chemotherapy, but are consequently responsive to toxicity [8]. Resistance to current chemotherapy for the treatment of NSCLC is one of the main obstacles to improve long-term efficacy for patients [9]. Although surgical resection is the most promising approach to cure NSCLC with improvements in conventional therapies over decades, the overall 5-year survival rate is modest, and further improvements in patient outcomes are needed [10,11].
MicroRNAs (miRs) are noncoding RNAs that can regulate the majority of protein-coding genes through either mRNA degradation or protein translation repression [12]. Previous studies have highlighted the crucial roles of miR-193a-3p and-5p, miR-143, and miR-148a in inhibiting migration and invasion of NSCLC cells [12][13][14]. miR-758 is involved in the pathogenesis of hemangiosarcoma, and is considered as a potential diagnostic biomarker of lupus nephritis in patients with systemic lupus erythematosus [15][16][17]. As highly abundant proteins, the high-mobility group box (HMGB) played both nuclear and extracellular roles in crucial biological processes [18]. As a member of HMGB subfamily, HMGB3 was recently reported to be an oncogene for leukemia, and is highly expressed in the progression phase of breast and gastric cancers [19]. HMGB3 has been shown to affect tumor initiation and progression, and silencing of HMGB3 expression inhibits proliferation, reduces migration in gastric cancer cells [20]. From the aforementioned, we speculate that there may be effects of miR-758 on proliferation and apoptosis of NSCLC cells via regulating the HGMB3 gene.

Ethical statement
The present study was performed with the approval of the Clinical Ethical Committee of Nanhai Hospital of Southern Medical University (People's Hospital of Nanhai District). All subjects signed informed consents prior to the study. All procedures were strictly conducted in accordance with the code of ethics.

Study subjects
A total of 50 NSCLC tissues and 50 adjacent tissues were obtained from NSCLC patients who underwent thoracic surgery in Nanhai Hospital of Southern Medical University (People's Hospital of Nanhai District) from January 2015 to January 2016. No patient underwent chemotherapy, radiotherapy, or other anti-cancer therapies before the surgery. All patients underwent surgical treatment with full medical history and follow-up information, and were diagnosed as primary NSCLC by pathological examination. The histological type and clinical pathological staging of the tumor were determined based on the lung and lung membrane tumors and Tumor Node Metastasis (TNM) staging criteria of the anticancer Alliance of World Health Organization (WHO) in 1997 [21]. Amongst them, there were 21 cases in clinical stage I, 17 cases in stage II, and 12 cases in stages III and IV; there were 20 cases of adenocarcinoma, 21 cases of squamous cell carcinoma, and 9 cases of poorly differentiated lung cancer categorized from pathological classification. The adjacent tissues were collected from at least 5 cm proximity from the NSCLC tissues, and identified with no tumor cell infiltration by Hematoxylin-Eosin (HE) staining. The NSCLC tissues and adjacent tissues were preserved in frozen tubes and stored in liquid nitrogen tanks.

Transient transfection
A549 cell line was selected and allocated into five groups: control (without transfection), miR-758 mimic (transfected with overexpressed miR-758), miR-758 mimic-negative control (NC) (transfected with miR-758 mimic NC), miR-758 inhibitor (transfected with inhibited miR-758), and miR-758 inhibitor-NC (transfected with miR-758 inhibitor NC) groups. All oligonucleotide sequences were synthesized by Shanghai GenePharma Co., Ltd. (Shanghai, China) ( Table  1). Twenty-four hours before transfection, the A549 cells were placed in the plate and incubated routinely. One hour before transfection, the original culture medium in each well was replaced with 2 ml of RPMI-1640 culture medium. The transfection mixture was prepared according to the instructions on the Lipofectamine 2000 kit (Invitrogen Inc., Carlsbad, CA, U.S.A.). The cells in the control group were only added with serum-free medium without penicillin/streptomycin medium; while the other four groups were added with serum-free and double antibody-free medium containing corresponding oligonucleotide fragments (the final concentration was 300 pmol/well) wrapped  by liposomes (Invitrogen Inc., Carlsbad, CA, U.S.A.). The transfected cells were cultured for 4 h in serum-free culture medium, added with 10% FBS, and then incubated in a 5% CO 2 incubator at 37 • C.

Reverse transcription quantitative PCR
RNA of NSCLC tissues, adjacent tissues, and cells in the five groups were extracted manually by TRIzol (Invitrogen Inc., Carlsbad, CA, U.S.A.) according to the instructions of the miRNeasy Mini kit (Qiagen Company, Hilden, Germany). The concentration and purity of the extracted RNA was detected using NanoDrop2000 (Thermo Fisher Scientific, Carlsbad, CA, U.S.A.), and the extracted RNA was stored at −80 • C for further usage. PCR reaction primer sequences ( Table 2) was designed using the primer design software Primer 5.0 (Premier, Palo Alto, CA, U.S.A.), and synthesized by Shanghai GenePharma Co., Ltd. (Shanghai, China). The ABI PRISM 7500 RT-PCR System (ABI Company, Oyster Bay, N.Y., U.S.A.), and the SYBR Green I fluorescent reagent kit (Takara Holdings Inc., Kyoto, Japan) were applied for PCR. With U6/β-actin as the internal reference gene, the reliability of PCR results was evaluated by the solubility curve, and the cycle threshold (C T ) value (inflexions in a kinetic PCR amplification curve) was obtained. C t = C T (target genes) C T (internal reference gene), C t = C t (the NSCLC group) − C t (the control group). Relative expression of target genes was calculated according to 2 − C t [22].

Western blot analysis
The cell protein was extracted after which the protein concentration was determined according to the instructions of the BCA kit (Boster Biological Technology Co., Ltd, Wuhan, Hubei, China). The extracted protein was added with the loading buffer, and boiled at 95 • C for 10 min. Each well was added with 30 μg of loading buffer, and protein separation was conducted by 10% PAGE (Boster Biological Technology Co., Ltd, Wuhan, Hubei, China). After electrophoresis at 80-120 V, the protein was transferred on to a PVDF membrane (100 mV for 45-70 min). Next, the cell protein was sealed in 5% BSA for 1 h, added with corresponding primary antibody (1:2500, Abcam Inc., Cambridge, MA, U.S.A.) and incubated overnight at 4 • C. The membrane was washed three times with TBS tween (TBST) buffer, 5 min each time, added with corresponding secondary antibody, and incubated at room temperature for 1 h. After three membrane washes for 5 min each time, the chemiluminescent reagent was used to develop with β-actin as an internal reference. GEL DOC EZ IMAGER (Bio-Rad, Inc., Hercules, CA, U.S.A.) was used to develop and gray values of target bonds were analyzed by the ImageJ software (National Institutes of Health, Bethesda, MA, U.S.A.).

Dual luciferase reporter gene assay
The microRNA.org site was used to predict the binding sites of miR-758 on HMGB3 mRNA 3 -UTR, and the HMGB3 3 -UTR wild type (3 -UTR-wt) and mutant (3 -UTR-mut) luciferase reporter vector containing miR-758 binding sites were constructed, respectively. The 293T cells were inoculated in a 24-well plate, and miR-758 mimic was co-transfected with HMGB3 3 -UTR-wt or HMGB3 3 -UTR-mut report vector using Lipofectamine TM 2000, with miR-758 mimic-NC set as NC. After transfection for 48 h, the luciferase activity was determined by the dual luciferase detection kit according to the instructions, expressing as the ratio of firefly luciferase to Renilla luciferase. The experiment was repeated in triplicates.

MTT assay
After transfection, A549 cells were diluted to a certain concentration and paved at a 96-well plate at the density of 5 ×

Wound-healing assay
Cells were seeded in a six-well plate, transfected, and cultured at 37 • C. When the cells adhered to the bottom of the plate, thin scratches were created along the center of each well with a sterile pipette tip (the width of each scratch was constant). The images were photographed (0 h) using an inverted microscope (Olympus Optical Co., Ltd., Tokyo, Japan) and marks were made in the six-well plate in order to position the same field. After incubation at 37 • C for 24 h, the culture medium was removed, and the cells were washed three times with PBS to remove surrounding cellular debris. Serum-free medium was then added to the wells, and images were acquired (24 h). Six fields at fixed position were selected to calculate the number of cells transferred to the scratch area using a click counting software.

Transwell assay
Transwell chambers were coated with Matrigel (3.9 mg/ml, 60-80 μl) and incubated at 37 • C for 30 min until Matrigel solidified. The cells were added with pre-warmed medium and hydrated in a 37 • C incubator for 2 h. Each well in the plate was added with 0.5 ml of complete medium, followed by addition of 0.5 ml of cell suspension (5 × 10 4 cells/ml) to each Transwell chamber. After incubation at 37 • C for 24 h, cells on the upper surface of the Transwell chamber were removed. The chamber was washed with PBS, soaked in pre-cooled methanol for 30 min to fix cells at the bottom of the chamber, and then stained with 0.1% Crystal Violet for 10 min. The images were photographed using an inverted microscope (Olympus Optical Co., Ltd., Tokyo, Japan). The cells that had transferred to the bottom chamber were counted in six fields and analyzed using counting software.

Flow cytometry
After transfection for 24 h, the cells were detached with trypsin, centrifuged, collected, and washed with cold PBS. The cells were re-suspended in PBS containing calcium in order to prepare a single-cell suspension (1 × 10 6 cells/ml). The cell suspension (100 μl) was collected into the tube, mixed with 10 mg/ml propidium iodide (PI) and 10 mg/ml RNase A, and then incubated at 4 expressed as percentages and rates, and analyzed by chi-square test. A probability value of P<0.05 was indicative of a statistically significant difference.

miR-758 shows lower expression in NSCLC tissues than in the adjacent tissues
The miR-758 expression in NSCLC and adjacent tissues was determined by reverse transcription quantitative PCR (RT-qPCR), showing that amongst the 50 pairs of NSCLC tissues, the miR-758 expression in 36 pairs of tissues presented with a lower miR-758 expression than adjacent tissues, with 14 pairs with significantly lower than that the adjacent tissues (>two-fold decrease). Statistical analysis showed that the miR-758 expression in NSCLC was significantly lower compared with the expression in adjacent tissues (P<0.05) (Figure 1).

HMGB3 shows higher expression in NSCLC tissues and has opposite tendency with miR-758
The HMGB3 expression in NSCLC and adjacent tissues was measured by RT-qPCR, showing that amongst the 50 cases of NSCLC tissues, the mRNA level of HMGB3 in 34 cases of tissues presented with a higher expression than the adjacent tissues, along with 16 cases of significantly higher (>two-fold increase) than the adjacent tissues (P<0.05) (Figure 2A). The results of Western blot analysis showed that the protein level of HMGB3 in 30 cases of NSCLC tissues was higher than that in the adjacent tissues, along with 20 cases with a significantly higher HMGB3 protein level (>two-fold increase) than the adjacent tissues (P<0.05) ( Figure 2B,C). Therefore, a conclusion can be drawn that the expression of HMGB3 is higher in NSCLC, with miR-758 following an antagonist trend. In addition, the relationship between the expression of miR-758 and HMGB3 and the clinicopathological features of NSCLC is shown in Supplementary Table S1. The expression of miR-758 and HMGB3 was correlated with clinical stage and pathological type of NSCLC (all P<0.05), but not with parameters such as age, gender, and smoking state (all P>0.05).

miR-758 is lowly expressed and HMGB3 is highly expressed in A549 cell line
The expression of miR-758 and HMGB3 in NSCLC cell lines was tested by RT-qPCR and Western blot analysis. In comparison with normal human lung endothelial cell BEAS-2B, miR-758 expression was lower in adenocarcinoma of lung cell line A549 (P<0.05), while it was significantly lower in H1650, H1975, and H292 cell lines (all P<0.01). Compared with A549 cell line, it was observed that the expression of miR-758 in other lung cancer cell lines decreased significantly (all P<0.05) ( Figure 3A). The expression of HMGB3 showed an opposite trend. The mRNA level of HMGB3 was lowest in the normal lung endothelial cell BEAS-2B, while it was higher in A549 (both P<0.05), and increased significantly in H1650, H1975, and H292 cell lines (all P<0.01). In comparison with A549 cell line, the mRNA level of HMGB3 in other lung cancer cell lines increased significantly (all P<0.05) ( Figure 3B). Western blot analysis results showed that the protein level of HMGB3 was lowest in normal lung endothelial cell BEAS-2B, while the same was higher in A549 (both P<0.05), and increased significantly in H1650, H1975, and H292 cell lines (all

HMGB3 is the target gene of miR-758
Dual luciferase reporter gene assay was performed to explore the target relationship between miR-758 and HMGB3. The results showed that after co-transfection of HMGB3 3 -UTR-wt and miR-758 mimic 293T cells, compared with the NC group (co-transfected with HMGB3 3 -UTR-wt and miR-758 mimic-NC), luciferase activity significantly decreased (P<0.05), with no significant change in luciferase activity after co-transfection of HMGB3 3 -UTR-mut and miR-758 mimic (P<0.05) ( Figure 4A,B). In comparison with the control group, HMGB3 expression had no significant change in the miR-758 mimic-NC and miR-758 inhibitor-NC groups (P>0.05); HMGB3 expression was significantly decreased in the miR-758 mimic group (P<0.05), but increased significantly in the miR-758 inhibitor group (P<0.05) ( Figure 4C-E).

Overexpression of miR-758 inhibits proliferation of A549 cells
MTT assay was applied to detect the NSCLC cell viability. After transfection for 24, 48, and 72 h, the effect of miR-758 on proliferation of A549 cells was detected and the results showed that compared with the control group, no significant change was observed in the cell survival rate between the miR-758 mimic-NC and miR-758 inhibitor-NC groups (all P>0.05); with observable significant changes in the miR-758 mimic and miR-758 inhibitor groups (all P<0.05). In comparison with the control and miR-758 mimic-NC groups, cell proliferation was significantly inhibited in the miR-758 mimic group (P<0.05). Compared with the control and miR-758 inhibitor-NC groups, it was observed that cell proliferation was significantly promoted in the miR-758 inhibitor group (P<0.05) ( Figure 5 and Supplementary material). The results indicated that overexpression of miR-758 inhibited proliferation of A549 cells.

Overexpression of miR-758 inhibits migration and invasion of A549 cells
The migration and invasion ability of transfected A549 cells were detected by wound-healing and Transwell assays. The results of wound-healing assay showed that compared with the control group, no significant changes were observable in the migration ability in the miR-758 mimic-NC and miR-758 inhibitor-NC groups (all P>0.05); but significant changes were observed in the miR-758 mimic and miR-758 inhibitor groups (all P<0.05). In comparison with the control and miR-758 mimic-NC groups, the cell migration was significantly inhibited in the miR-758 mimic group (P<0.05). In comparison with the control and miR-758 inhibitor-NC groups, cell migration was significantly promoted in the miR-758 inhibitor group (P<0.05) ( Figure 6A,C). Transwell assay results demonstrated that the invasion ability of the mimic-NC and inhibitor-NC groups was not significantly different from that of the control group (both P>0.05), and the invasion ability of the miR-758 mimic and miR-758 inhibitor groups significantly changed compared with that of the control group (both P<0.05). The invasion ability of the miR-758 mimic group was significantly lower than that of the control and mimic-NC groups (P<0.05). In comparison with the control and inhibitor-NC groups, the miR-758 inhibitor group displayed increased significantly cell invasion rate (P<0.05) ( Figure 6B,D). The results indicated that overexpression of miR-758 inhibited the migration and invasion of A549 cells.

Overexpression of miR-758 prevents cells from entering the S phase from G 1 phase
After transfection for 24 h, the cell cycle of A549 in each group was detected by flow cytometry, and the results showed that compared with the control group, no significant changes were observed in the cell cycle in the miR-758  mimic-NC and miR-758 inhibitor-NC groups (all P>0.05). In comparison with the control and miR-758 mimic-NC groups, the cells were arrested in G 1 phase in the miR-758 mimic group, with the miR-758 mimic group presenting with more cells in G 1 phase but fewer cells in S and G 2 phases (all P<0.05). In comparison with the control and miR-758 inhibitor-NC groups, the miR-758 inhibitor group had fewer cells arrested in G 1 phase and substantially more cells in S and G 2 phases (P<0.05) ( Table 3). The results showed that overexpressed miR-758 prevented cells from entering the S phase from G 1 phase.

Overexpression of miR-758 promotes apoptosis of NSCLC cells
Flow cytometry was employed in order to detect the effect of miR-758 on NSCLC cell apoptosis. After transfection for 24 h, the apoptosis of A549 cells in each group was detected, and the results showed that compared with the control group, the cell apoptosis rate had no statistical difference in the miR-758 mimic-NC and miR-758 inhibitor-NC groups (all P>0.05). In comparison with the control and miR-758 mimic-NC groups, cell apoptosis rate was significantly increased in the miR-758 mimic group (P<0.05). Compared with the control and miR-758 inhibitor-NC groups, cell apoptosis rate was significantly decreased in the miR-758 inhibitor group (P<0.05) (Figure 7). The results demonstrated that overexpression of miR-758 promoted apoptosis of NSCLC cells.

Discussion
NSCLC is the main cause of cancer-related mortality, with high prevalence of 85% amongst all types of lung cancer cases over the world each year [23,24]. miRs have been proven to be potential markers in NSCLC due to their positive roles in oncogenesis [25]. Previous studies have identified that miRs could regulate the proliferation, migration, and cancer pathogenesis of several cancer types, including NSCLC [26,27]. In this study, the researchers explored roles of miR-758 in proliferation and apoptosis of NSCLC cells via regulating the HMGB3 gene. Finally, our experimental results indicated that overexpression of miR-758 suppressed proliferation and promoted the apoptosis of NSCLC cells by negatively modulating the HMGB3.
Initially, a significantly lower miR-758 expression was observed in NSCLC tissues in contrast with the adjacent tissues, with a high HMGB3 expression in NSCLC tissues than the adjacent tissues. Our results were in consistency with the results of a study conducted by Ma et al. [28], highlighting a remarkably low expression of miR-361-5p in NSCLC tissues than in adjacent tissues. The relative expression of miR-503 was significantly lower in NSCLC tissues, and was associated with various cellular processes such as lymphatic invasion, distant metastasis, TNM stage and tumor grade; additionally, low miR-503 expression is an independent prognostic factor for the overall survival in NSCLC [29]. HMGB3, an X-linked member, belongs to the high-mobility group family and functions as a common protein in the mediation of nucleic acid-mediated innate immune responses [30]. HMGB3 has been observed to contribute to numerous tumor progressions [31]. A high expression of HMGB3 was observed in the progression phase of breast cancer [32]. A study conducted by Tang et al. [33] identified HMGB3 as one of the evidently detectable biomarkers in peripheral blood in lung cancer. Song et al. [34] found a notably higher level of HMGB3 in NSCLC tissues than in adjacent tissues, and overexpression of HMGB3 shared close association with advanced tumor progression as well as unsatisfactory clinical outcomes of patients with NSCLC.
Besides, HMGB3 expression was significantly decreased in the miR-758 mimic group whereas it was increased significantly in the miR-758 inhibitor group, indicating that miR-758 negatively regulated HMGB3 expression. The results of dual luciferase reporter gene assay were in consistency with the aforementioned. Similarly, previous studies demonstrated that overexpressions of miR-181b and miR-26a inhibited HMGB1 expression [35,36]. A former study highlighted the association of high-mobility group proteins with cancer progression [37]. Likewise, a recent study found HMGB1 as the target gene of miR-142-3p, negatively regulated by miR-142-3p [38]. Zhu et al. [39] found that HMGB1 stimulated drug resistance in NSCLC cells along with migration in human NSCLC cells [40]. A correlation between overexpression of HMGB1 with the poor survival rate for NSCLC patients was evident [41]. A prior study speculated a relationship between HMGB3 and miR overexpression, namely up-regulation of miR-513b inhibited cell proliferation, migration, and promoted apoptosis by targetting the HMGB3 protein in gastric cancer [19]. The finding showed that overexpression of miR-758 inhibited HMGB3 expression and may be a novel target for NSCLC.
Decisively, in the miR-758-mimic group, proliferation, migration, and invasion were inhibited along with stimulated apoptosis in NSCLC, whereas opposite results were observed in the miR-758 inhibitor group. In addition, miR-758-3p overexpression also displayed significant suppressive effect on proliferation, migration, and invasion of hepatocellular carcinoma cells [42]. It has been reported that up-regulation of miR-27b suppresses the proliferation, migration, and invasion of A549 NSCLC cells [43]. A relationship has been observed between overexpression of miR-7 and repressed proliferation, migration, tumorigenicity, and increased apoptosis in NSCLC cells, which was in consistency with our results [44]. Similarly, overexpression of miR-3127-5p significantly reduced proliferation, migration, and motility of NSCLC cells in vitro and in vivo [45]. The results demonstrated that overexpression of miR-758 inhibited proliferation, migration, and invasion but increased apoptosis of NSCLC cells via inhibiting the HMGB3 gene.
To conclude, our data speculated that overexpression of miR-758 inhibits proliferation, migration, invasion, and cell cycle, and stimulates apoptosis of NSCLC cells by negatively regulating the HMGB3. However, due to contribution of various other affecting factors of the miR-758 and HMGB3 gene, further studies are needed to explore the mechanism of miR-758 and HMGB3 gene and find more effective therapeutic regimens for NSCLC patients.