Abstract

The present study aims to investigate the mechanism of miR-384 in non-small cell lung cancer (NSCLC) cell apoptosis and autophagy by regulating Collagen α-1(X) chain (COL10A1). Bioinformatics methods were applied to evaluate potential miRNAs and genes that might correlate with NSCLC. Tumor tissues and adjacent tissues from 104 NSCLC patients were collected and human NSCLC A549 cell line was selected for subsequent experiments. A549 cells were treated with miR-384 mimic, miR-384 inhibitor, or knockdown of COL10A1. Quantitative real-time PCR (qRT-PCR) and Western blotting were utilized to detect the levels of miR-384, COL10A, Survivin, Bcl-2, Bax, Bcl-xl, Beclin 1, and LC3 in tissues and cells. A series of biological assays including MTT assay, Annexin V-FITC/PI (propidium iodide) staining, immunofluorescence, monodansylcadaverine (MDC) staining were conducted to investigate the effects of miR-384 and COL10A1 on NSCLC cells. Tumorigenicity assay for nude rats was applied. Results obtained from the present study indicated that miR-384 down-regulated COL10A1 by targetting it. Compared with adjacent tissues, miR-384 expression was obviously reduced while COL10A1 expression was significantly enhanced in NSCLC tissues (all P<0.05). Outcomes in vivo and in vitro suggested that cell proliferation and tumorigenicity were inhibited while cell apoptosis and autophagy were induced in NSCLC cells treated with up-regulation of miR-384 or silence of COL10A1. In miR-384 inhibitor group, cell proliferation was improved, while cell apoptosis was reduced and cell autophagy was decreased whereas tumorigenicity of cells was strengthened. Based on the findings of our study, it was established that miR-384 could down-regulate COL10A1 levels, subsequently inhibiting cell proliferation and promoting cell apoptosis and autophagy in NSCLC cells.

Introduction

Non-small cell lung cancer (NSCLC) featured by high incidence is one of the most malignant cancer in China and there is lack of major advancements in this treatment [1]. NSCLC includes various types such as adenocarcinoma, squamous cell carcinoma, and large cell carcinoma [2]. NSCLC patients are at high risk of poor prognosis and the main treatment method is surgical therapy followed by radiotherapy and chemotherapy [3]. Since approximately 70–80% NSCLC patients do not show noticeable clinical symptoms at early stage, they had lost the opportunity of surgical therapy [4]. At the same time, have there are too many adverse side effects in radiotherapy as well as chemotherapy, which can cause damage to patients’ health [5]. Thus, novel therapeutic strategies are needed to improve clinical effects in NSCLC patients, which makes molecule-targetted therapy a widely investigated therapeutic approach now.

miRNAs are short and non-coding RNAs and normally contain 21–23 nts [6]. Accumulating evidence indicates that various types of cancers were often accompanied by abnormal expression of miRNAs. Therefore, miRNAs are of great significance to be biomarkers for diagnosis and treatment of cancer [7,8]. In addition, increasing evidence demonstrates that abnormal expression of miRNAs is also associated with activation of oncogene and mutation, inactivation and abnormal expression of anti-oncogene, which is the main cause of cancer [9]. miR-384 has been reported to be associated with the progressions of many types of tumors, ranging from pancreatic cancer to renal cell carcinoma [10,11]. Another study searched by Fan et al. [12] confirms that miR-384 is significantly down-regulated in NSCLC tissues and cells, indicating miR-384 might be an important therapeutic target for NSCLC. Collagen α-1(X) chain (COL10A1), which encodes the α chain of type X collagen, is confirmed to be a member of the collagen family [13]. Normally, COL10A1 is highly expressed in several types of human tumors, while it exists at a lower level in normal tissues [14,15]. Studies on COL10A1 and colorectal cancer have proved that high expression of COL10A1 is an independent risk factor for prognosis and overall survival rates in patients with colorectal cancer [16]. In this study, we identified COL10A1 as an NSCLC-associated gene by screening it through the bioinformatics databases and miR-384 down-regulated COL10A1 by targetting it. Thus in the present study, we are aiming to investigate the mechanism of miR-384 targetting COL10A1 in NSCLC.

Materials and methods

Bioinformatics prediction of molecular mechanism in NSCLC

The gene expression chips were searched in the Gene Expression Omnibus (GEO) (http://www.ncbi.nlm.nih.gov/geo) database using keywords ‘lung cancer’. Amongst the various associated chips, GSE19804 and GSE27262 were chosen for screening differentiation gene expression. Differentially expressed genes of lung cancer were identified with the limma package of R language (http://master.bioconductor.org/packages/release/bioc/htmL/limma.htmL) [17]. The adjusted P-value (p.adj) was set as 0.01 and the absolute value of log2 fold change (log2FC) was set as 2 to analyze the significantly differentially expressed gene which would been shown on a heat map. RNA22 (https://cm.jefferson.edu/rna22/), TargetScan (http://www.targetscan.org/vert_71/), and miRNApath (http://lgmb.fmrp.usp.br/mirnapath/index.php) were used as tools to predict the miRNA–mRNA interaction and the regulation of miRNAs to their target genes. Venny2.1.0 (http://bioinfogp.cnb.csic.es/tools/venny/index.html), an online analyzing tool to draw Venn diagrams, was used to display common miRNAs identified by miRNA tools.

Identification of potential target genes of miRNA and experimental validation

miR-384 target genes were further fished out by using the online miRNA prediction tools (https://cm.jefferson.edu/rna22/). COL10A1 was identified as a potential target gene of miR-384 linked to NSCLC, which was validated by dual luciferase reporter assay. The gene fragments of COL10A1 3′-UTR in wild-type (WT) and mutation (MUT) type were amplified and inserted into multiple clone sites of pmirGLO vector, respectively (E1330, Promega Corporation, U.S.A.). The Renilla luciferase reporter plasmid (E2241, Promega Corporation, U.S.A.) was set as the internal reference to adjust the differences of cell numbers and efficiency of trasfection. pmirGLO vector containing COL10A1-3′UTR-WT (COL10A1-3′UTR-MUT) and miR-384 mimic (scrambled negative control (NC)) were co-transfected into HEK-293T cells (CL-0005, Procell, China). Forty-eight hours after transfection, cells were labeled with Dual-Luciferase Reporter Assay System Kit (Promega, U.S.A.) and fluorescence intensity was measured using fluorescence microscope (XSP-BM22AY, Shanghai Optical Instrument Factory, China).

Study subjects

From January 2015 to January 2018, 104 patients (with a mean age of 57.2 ± 14.2 years, 65 men and 39 women) clinically and pathologically diagnosed with NSCLC in our hospital were enrolled in this research. Inclusion criteria were as follows: (i) patient without history of malignant tumors; (ii) patient without receiving any treatments such as chemotherapy, radiotherapy, or other treatments prior to the operation described in the present study; (iii) patient with complete clinicopathological and follow-up data [18]. In total, NSCLC tumors were moderately differentiated and well-differentiated in 46 patients and poorly differentiated in 58 patients. According to Tumor Node Metastasis (TNM) staging standard [19], 65 cases were classified as stage I while 39 cases were diagnosed at stage II. In the above patients, lymphatic metastasis was found in 46 cases which did not exist in other 58 cases. Tumor tissues and adjacent normal tissues (3–5 cm from the edge of cancer tissues) were collected from NSCLC patients. All tissue samples were treated with liquid nitrogen at −196°C. We obtained each patient’s informed consent and the Ethics Committee of Tongren Hospital approved this research.

Cell culture and selection for high expression of COL10A1

BEAS-2B (normal lung epithelial cell line), A549 (lung adenocarcinoma cell line), and GLC82 and MES-1 and LTEP-s (lung squamous cell carcinoma cell lines) were purchased from American Type Culture Collection (ATCC, U.S.A.). BEAS-2B, A549, GLC82, MES-1, and LTEP-s cells were cultured in RPMI 1640 culture medium (GNM-11879, Shanghai Jing Ke Chemical Technology Co., Ltd, China) supplemented with 10% FBS (HyClone, Logan, Utah, U.S.A.), along with 100 U/ml penicillin and 100 mg/ml streptomycin. The cells were incubated at 37°C in a constant-temperature incubator with 5% CO2. Fresh culture medium was substituted every 1 or 2 days. Quantitative real-time PCR (qRT-PCR) and Western blotting were performed to choose cell line with the highest COL10A1 expression for further experiments.

Construction of recombinant plasmid containing COL10A1 siRNA

The COL10A1 siRNA (siRNA1: 5′-CCAAATGCCCACAGGCATA-3′; siRNA2: 5′-TCTTCATTCCCTACACCAT-3′; siRNA3: 5′-CCAAGACACAGTTCTTCAT-3′) and NC sequence (5′-CCACACATTGATTCGACAT-3′) were designed using BLOCK-iT™ RNAi Designer (http://maidesigner.thermofisher.com/maiexpress) and synthesized by Thermo Fisher Scientific Co., Ltd. Next, the synthesized sequences were inserted into pcDNA3.1(+) (VPI0001, Invitrogen, U.S.A.) which was cut by Hind III and XHo I restriction endonuclease and T4 ligase was used for ligation between pcDNA3.1 and goal sequences. And the recombinant plasmids were transformed into competent Escherichia coli DH5α (D9052, Takara, Japan). The resistant colony was picked and cloned, DNA of which was extracted via Genomic DNA Mini Preparation Kit (D0063, Beyotime, China) and identified using enzyme digestion and PCR. After that recombinant plasmids were extracted by PicoPure™ DNA Extraction Kit (KIT0103, Thermo Fisher, U.S.A.) and preserved at −20°C. Western blotting was performed for selection for siRNA which had the best interference efficiency.

Cell transfection and grouping

NSCLC cells were assigned the following groups: blank group (without any transfection), NC group (transfected with unrelated sequence), miR-384 mimic group (transfected with miR-384 mimic), miR-384 inhibitor group (transfected with miR-384 inhibitor), siRNA-COL10A1 group (transfected with recombinant plasmid containing siRNA-COL10A1 sequence), miR-384 inhibitor+siRNA-COL10A1 group (transfected with miR-384 inhibitor and recombinant plasmid containing siRNA-COL10A1 sequence). NSCLC cells were transfected referring to Lipofectamine 2000 instructions (11668027, Thermo Fisher Scientific Inc., Waltham, MA, U.S.A.). During transfection, 6 µl Lipofectamine 2000 was diluted in 200 µl serum-free Opti-MEM (31985070, Gibco, Gaitherburg, MD, U.S.A.) and 2 μg plasmids were diluted in serum-free Opti-MEM (100 µl). Above two solutions were mixed and incubated at room temperature for 30 min. And then the mixed solution was added into culture wells (106 cells/well for 24-well plates). One milliliter of serum-free medium Opti-MEM was gently added into each well and the cells were then incubated in 5% CO2 incubator at 37°C for 48 h. Transfection of miR-384 mimic (20 pmol for 106 cells/well), miR-384 inhibitor (22 pmol for 106 cells/well), NC (20 pmol for 106 cells/well), and miR-384 inhibitor+siRNA-COL10A1 (106 cells were treated with 22 pmol miR-384 inhibitor after 1 h for 2 μg siRNA plasmids transfection) which were designed and synthesized by The Beijing Genomics Institute followed the above methods.

qRT-PCR

Total RNA from cells and NSCLC and adjacent tissues were extracted using TRIzol reagent (16096020, Thermo Fisher Scientific Inc., Waltham, MA, U.S.A.). UV spectroscopy (UV1901, Aoxi Scientific Instrument Ltd., Shanghai, China) was adopted to examine the purity and concentration of RNA in the tissues and cells. The concentration of all the samples (A260/A280 = 1.8–2.0) was then adjusted to 100 ng/μl. Three microliters of extracted RNA was reverse transcribed into cDNA using cDNA Reverse Transcriptase Kits (18090010, Thermo Scientific, U.S.A.). Reaction system: 10× RT buffer 2.0 μl, 25× dNTP mix (100 mM) 0.8, 10× RT Random Primers 2.0 μl, MultiScribe™ Reverse Transcriptase 1.0 μl, RNase inhibitor 1.0 μl, Nuclease-free H2O2 3.2 μl. Reaction conditions: step 1: 25°C for 10 min; step 2: 37°C for 120 min; step 3: 85°C for 5 min; step 4: preservation at 4°C. qRT-PCR was conducted using the miScript SYBR Green PCR Kit (Qiagen Company, Valencia, CA). The primer sequences are shown in Table 1. Twenty-five microliters of reaction systems were employed as follows: 300 ng cDNA, 1× PCR buffer, 200 μmol/l dNTPs, forward and reverse primers 80 pmol/l, and 0.5 U Tag polymerase. And the reaction conditions were as follows: 95°C for 10 min, 95°C for 15 s, and 60°C for 1 min. The fluorescence signal was collected using PCR instrument and 45 cycles were conducted. U6 was set as the internal reference of miR-386 while glyceraldehyde phosphate dehydrogenase (GAPDH) was set as the internal reference of other genes. Finally the 2−ΔΔCt method was adopted to measure the relative mRNA expression in tissues and cells [20].

Table 1
Primer sequences of qRT-PCR
Primer Sequence 
COL10A1 F: 5′-TCTGTGAGCTCCATGATTGC-3′ 
 R: 5′-GCAGCATTACGACCCAAGATC-3′ 
miR-384 F: 5′-TGTTAAATCAGGAATTTTAA-3′ 
 R: 5′-TGTTACAGGCATTATGAA-3′ 
BAX F: 5′-CATATAACCCCGTCAACGCAG-3′ 
 R: 5′-GCAGCCGCCACAAACATAC-3′ 
LC3 F: 5′-GGTTTCCCGTCACCAATTTTCC-3′ 
 R: 5′-TGTGGTTTCCAACGTAGAGGA-3′ 
Bcl-xl F: 5′-GGAATICATGTCTCAGAGCAACCGG-3′ 
 R: 5′-CTGATCXGCGGTTGAAGCGTTCCTG-3′ 
Bcl-2 F: 5′-GTCTTCGCTGCGGAGATCAT-3′ 
 R: 5′-CATTCCGATATACGCTGGGAC-3′ 
Beclin1 F: 5′-AAGACAGAGCGATGGTAG-3′ 
 R: 5′-CTGGGCTGTGGTAAGTAA-3′ 
Survivin F: 5′-GCA TGG GTG CCC CGA CGT TG-3′ 
 R: 5′-GCT CCG GCC AGA GGC CTC AA-3′ 
GAPDH F: 5′-CCATGTTCGTCATGGGTGTGAACCA-3′ 
 R: 5′-GCCAGTAGAGGCAGGGATGATGTTC-3′ 
U6 F: 5′AAAGCAAATCATCGGACGACC-3′ 
 R: 5′-GTACAACACATTGTTTCCTCGGA-3′ 
Primer Sequence 
COL10A1 F: 5′-TCTGTGAGCTCCATGATTGC-3′ 
 R: 5′-GCAGCATTACGACCCAAGATC-3′ 
miR-384 F: 5′-TGTTAAATCAGGAATTTTAA-3′ 
 R: 5′-TGTTACAGGCATTATGAA-3′ 
BAX F: 5′-CATATAACCCCGTCAACGCAG-3′ 
 R: 5′-GCAGCCGCCACAAACATAC-3′ 
LC3 F: 5′-GGTTTCCCGTCACCAATTTTCC-3′ 
 R: 5′-TGTGGTTTCCAACGTAGAGGA-3′ 
Bcl-xl F: 5′-GGAATICATGTCTCAGAGCAACCGG-3′ 
 R: 5′-CTGATCXGCGGTTGAAGCGTTCCTG-3′ 
Bcl-2 F: 5′-GTCTTCGCTGCGGAGATCAT-3′ 
 R: 5′-CATTCCGATATACGCTGGGAC-3′ 
Beclin1 F: 5′-AAGACAGAGCGATGGTAG-3′ 
 R: 5′-CTGGGCTGTGGTAAGTAA-3′ 
Survivin F: 5′-GCA TGG GTG CCC CGA CGT TG-3′ 
 R: 5′-GCT CCG GCC AGA GGC CTC AA-3′ 
GAPDH F: 5′-CCATGTTCGTCATGGGTGTGAACCA-3′ 
 R: 5′-GCCAGTAGAGGCAGGGATGATGTTC-3′ 
U6 F: 5′AAAGCAAATCATCGGACGACC-3′ 
 R: 5′-GTACAACACATTGTTTCCTCGGA-3′ 

Western blot

RIPA lysis buffer (R0010, Beijing Solarbio Life Sciences Co., Ltd, Beijing, China) was applied for total protein extraction of NSCLC tissues and cells and the detection of total protein concentration was detected following the instructions of BCA protein assay kit (P0009, Beyotime Biotechnology Co., Ltd., Shanghai, China). Then the proteins was adjusted to suitable concentrations using deionized water, separated by SDS/PAGE, transferred on to PVDF (ISEQ00010, Beijing Solarbio Life Sciences Co., Ltd., Beijing, China) membrane. The membrane was sealed by 5% skimmed milk overnight at 4°C. Next day, primary antibodies were added for 2 h incubation at 37°C: rabbit-anti-human COL10A1 (1:1000, ab182536, Abcam Inc., Cambridge, MA, U.K.), Bax (1:10000, ab32503, Abcam Inc., Cambridge, MA, U.K.), LC3B (LC3II, LC3I) (1:3000, ab51520, Abcam Inc., Cambridge, MA, U.K.), Survivin (1:5000, ab76424, Abcam Inc., Cambridge, MA, U.K.), Beclin 1 (1:2000, ab207612, Abcam Inc., Cambridge, MA, U.K.), Bcl-2 (1:1000, Abcam Inc., Cambridge, MA, U.K.), Bcl-xl (1:1000, ab32370, Abcam Inc., Cambridge, MA, U.K.), and GAPDH (1:2500, ab9485, Abcam Inc., Cambridge, MA, U.K.). And then the membrane was incubated with horseradish peroxidase (HRP)-labeled secondary antibody IgG (1:2500, ab99697, Abcam Inc., Cambridge, MA, U.K.) at 4°C overnight. Electrochemiluminescence (ECL, 808-25, Biomiga, U.S.A.) was used for 15-min color reaction at 37°C. GAPDH was set as the internal reference. And gray value of protein bands was analyzed using ImageJ software (National Institutes of Health, U.S.A.).

MTT assay

Forty-eight hours after transfection, cells were digested using 0.25% trypsin for 1 min and resuspended with RPMI 1640 medium containing 10% FBS. Suspended cells were seeded on a 96-well plate with the cell density adjusted to 1 × 104 cells/ml (100 μl/well) and fostered at 37°C in a constant-temperature incubator with 5% CO2 for 2 h. To each well, 20 μl MTT (5 mg/ml, GD-Y1317, Shanghai Guduo Biotech Co., Ltd.) was added, after which the plate was put in the incubator for 3 h. After centrifugation at 1000 rpm for 5 min, the supernatant was discarded. Next, each well was added with 150 μl DMSO (D5879-100ML, Sigma, U.S.A.) shocked for 10 min to dissolve crystals. The OD value of each well was measured at 490 nm by ELISA reader (BIOBASE-EL10A, Ji’nan Biobase Medical Equipment Co., Ltd., China). Each experiment was repeated three times and the cell viability curve was plotted with time as x-axis and OD value as y-axis.

Annexin-V-FITC/propidium iodide staining

Forty-eight hours after transfection, cells were digested using 0.25% trypsin–EDTA for 1 min and resuspended with proper PBS. Two hundred microliters of cell suspension was centrifuged at 1000 rpm for 5 min and then supernatant was discarded. Next, cells were washed with PBS three times for 5 min each. Subsequently, 195 μl Annexin V-FITC binding buffer was utilized for suspending cells, after which 5 μl Annexin-V-FITC (20 μg/ml) was gently added into above system for 15 min incubation on ice. The mixture was transferred into flow detection tube containing 400 μl of PBS. Ten microliters of propidium iodide (PI, 50 μg/ml) was added into tube for 2 min incubation on ice in the dark. At last, the cell apoptosis was measured using flow cytometry (FACS Calibur, BD Biosciences, U.S.A.) within 30 min.

Immunofluorescence

Coverslips were immersed in cells’ medium to let cells grow on it. And then they were washed with PBS for three times and 5 min each time. Paraformaldehyde (4%) was applied to fix cells on coverslips for 15 min at room temperature. Coverslips were washed with PBS for three times and 3 min each time. And then coverslips were incubated with PBS containing 0.5% Triton X-100 for 5 min. Next, the cells were sealed using 5% skimmed milk for 1 h, after which LC3II antibody was added (diluted to 1 µg/ml, ab48394, Abcam Inc., Cambridge, MA, U.K.). After incubation overnight at 4°C, the coverslips were washed with PBS once and incubated with Alexa Fluor 488 labeled donkey anti-rabbit antibody IgG (1:1000, ab175475, Abcam Inc., Cambridge, MA, U.K.) at room temperature for 1 h. Next, the coverslips were washed with PBS and stained using Hoechst 33342 (C1029, Beyotime Biotechnology Co., Ltd., Shanghai, China). After washing with PBS, coverslips were sealed with ProLong™ Live Antifade Reagent (P36974, Thermo Fisher, U.S.A.). Finally, the cells were observed under fluorescence microscope (WSB-1200A, Guangzhou Weiscope Optical Instrument Factory, Guangzhou, China).

Monodansylcadaverine assay

NSCLC cells in logarithmic growth phase were collected through centrifugation, digested with 0.25% trypsin for 1 min, and seeded on a six-well plate with concentration at 3 × 104 cells/well. After cells growing adherently for 24 h, the plate was further incubated in 5% CO2 incubator for 30 min at 37°C in darkness. After that, cells were centrifuged and resuspended with PBS. Ninety microliter of cell (106 cells/ml) suspensions were collected and added with 10 μl monodansylcadaverine (MDC) (OKWB00363, Beijing Aviva Biotech Co., Ltd. Beijing, China) staining solution. After staining for 30 min at room temperature, the cell suspension was centrifuged at 2000 rpm for 5 min, after which cells were collected and washed with PBS for three times. And then the suspension was dropped on to slide glass, covered with coverslips, which was observed by using fluorescence microscope (F36914, Thermo Fisher, U.S.A.) with 355-nm emission filter and 512-nm block filter.

Tumorigenicity assay in nude rats

Subsequently, thirty 4-week old specific-pathogen free (SPF) grade nude rats (provided by Medical Laboratory Animal Center of Zhongshan University) weighing 22–27g, male and female in half. Rats were fed in SPF level clean room and had access to food and water. The transfected NSCLC cells in logarithmic growth phase were adjusted to 1 × 107 cells/ml with RPMI 1640 medium, and 200 μl cells were injected into nude rats through the right-side armpit skin (n=5 per group). The rats of each group were then fed in the same environment, and the tumors were observed and recorded once a week from first week. The volumes of the tumor = length × width2/2. Rats were killed in fifth week and tumors were taken out and photographed. Animal use and all experimental procedures were carried out in accordance with Guidelines for the management and use of experimental animals and approved by Experimental Animal Ethics Committee of Tongren Hospital in China.

Statistical analysis

SPSS 22.0 software (IBM Corp., Armonk, NY, U.S.A.) was applied for data analyses. Measurement data were expressed as mean ± S.D. (x ± s). Comparisons between two groups were conducted using a Student’s t test. Comparisons amongst multiple groups were examined by one-way ANOVA for variation analysis and significance analysis. The normality test was conducted using the Kolmogorov–Smirnov method. Normally distributed data amongst multiple groups were compared using one-way ANOVA with Tukey’s post-hoc test, while skewed data distributed amongst multiple groups were analyzed using Dunn’s multiple comparison for post-hoc tests following Kruskal–Wallis testing. P<0.05 was indicative of statistical significance.

Results

Bioinformatics prediction of NSCLC-associated miRNA and target gene

A systematic multistep approach combining miRNA and mRNA expression profiles and bioinformatics analysis was adopted to identify the NSCLC-specific interactions between miRNAs and mRNAs. According to the information screened from analysis of gene chip GSE19804 and GSE27262, we identified a total of 120 and 311 differentially expressed genes via setting |logFC| > 2.0 and adj.P.Val < 0.01 as threshold values. The top ten differential expression genes were selected and expressions of these genes were illustrated in the heat maps (Figures 1 and 2). Compared with adjacent normal tissues, COL10A1 was extremely highly expressed in lung cancer tissues in GSE19804 and GSE27262 microarrays with a significant difference (all P<0.05) (Figure 3). COL10A1 showed an up-regulation in colorectal cancer and its high expression was correlated with poor prognosis of colorectal cancer [13,16], while little research has been engaged in role of COL10A1 in NSCLC. Therefore, our study focussed on the impact of COL10A1 in NSCLC and possible molecular mechanism of regulating COL10A1. To predict potential miRNAs participating in modulating COL10A1 expression, three miRNA prediction tools, TargetScan, RNA22, and miRNApath, were applied and 314, 1645, and 23 miRNAs predicted to combine with COL10A1 were identified, respectively. Amongst these results obtained by three tools, hsa-miR-384 and hsa-miR-581 were common miRNAs in Venn diagram (Figure 4). There was little study in regard to miR-581 in cancer research, while miR-384 has been proved to inhibit cell proliferation and migration in several cancers [10,21,22]. Considering that the role of miR-384 in NSCLC still remains unknown, we put forward the hypothesis that miR-384 is likely modulating the NSCLC through interacting with COL10A1.

Differentially expressed genes in GSE19804

Figure 1
Differentially expressed genes in GSE19804

The heat map showing ten differentially expressed genes on GSE19804 chip. The x-axis indicates the sample number, the y-axis shows the differentially expressed genes, and the right upper histogram is the color gradation. Each rectangle in the figure corresponds to an individual sample. The red and blue colors represent relatively high and low fold-change of expression, respectively.

Figure 1
Differentially expressed genes in GSE19804

The heat map showing ten differentially expressed genes on GSE19804 chip. The x-axis indicates the sample number, the y-axis shows the differentially expressed genes, and the right upper histogram is the color gradation. Each rectangle in the figure corresponds to an individual sample. The red and blue colors represent relatively high and low fold-change of expression, respectively.

Differentially expressed genes in GSE2726

Figure 2
Differentially expressed genes in GSE2726

The heat map showing ten differentially expressed gene on GSE2726 chip. The abscissa indicates the sample number, the ordinate indicates the difference gene, and the right upper histogram is the color gradation. Each rectangle in the figure corresponds to one sample expression. The red and blue colors represent relatively high and low fold-change of expression, respectively.

Figure 2
Differentially expressed genes in GSE2726

The heat map showing ten differentially expressed gene on GSE2726 chip. The abscissa indicates the sample number, the ordinate indicates the difference gene, and the right upper histogram is the color gradation. Each rectangle in the figure corresponds to one sample expression. The red and blue colors represent relatively high and low fold-change of expression, respectively.

Expression of COL10A1 in microarrays

Figure 3
Expression of COL10A1 in microarrays

(A,B) Expression of COL10A1 in normal adjacent and lung cancer tissue indicated by GSE19804 and GSE2726 chip.

Figure 3
Expression of COL10A1 in microarrays

(A,B) Expression of COL10A1 in normal adjacent and lung cancer tissue indicated by GSE19804 and GSE2726 chip.

Prediction for upstream miRNA of COL10A1

Figure 4
Prediction for upstream miRNA of COL10A1

Comparison of TargetScan, RNA22, and miRNApath on COL10A1 upstream miRNA, hsa-miR-384, and hsa-miR-581 were identified as the common miRNAs.

Figure 4
Prediction for upstream miRNA of COL10A1

Comparison of TargetScan, RNA22, and miRNApath on COL10A1 upstream miRNA, hsa-miR-384, and hsa-miR-581 were identified as the common miRNAs.

miR-384 down-regulated COL10A1 expression

The target gene of miR-384 was determined via RNA22 prediction tool (https://cm.jefferson.edu/rna22/). And specific-binding sites were shown between miR-384 and COL10A1 (Figure 5A). This was further validated by dual luciferase reporter assay. As indicated in Figure 5B, a significant decrease in luciferase intensity was observed in HEK-293T cells co-transfected with miR-384 mimic and COL10A1-3′UTR-WT plasmid compared with cells co-transfected with NC and COL10A1-3′UTR-WT sequence (all P<0.05). No significant difference was observed amongst other groups (all P>0.05). Such results provided direct evidence that miR-384 combined with COL10A1-3′UTR.

Verification for combination of miR-384 and COL10A1 by luciferase reporter gene assay

Figure 5
Verification for combination of miR-384 and COL10A1 by luciferase reporter gene assay

(A) Binding site of miR-384 on COL10A1. (B) The luciferase activity was detected after transfection for 48 h. Data are shown as mean ± S.D. and experiments were repeated three times. *, P<0.05 when compared with NC group.

Figure 5
Verification for combination of miR-384 and COL10A1 by luciferase reporter gene assay

(A) Binding site of miR-384 on COL10A1. (B) The luciferase activity was detected after transfection for 48 h. Data are shown as mean ± S.D. and experiments were repeated three times. *, P<0.05 when compared with NC group.

miR-384 level was decreased while COL10A1 level was improved in NSCLC tissues

The expressions of miR-384 and COL10A1 in tumor and adjacent tissues were quantitated utilizing qRT-PCR and Western blot. Compared with adjacent tissues, miR-384 expression in NSCLC tissues was obviously down-regulated, while the mRNA and protein expressions of COL10A1 were significantly induced (all P<0.05) (Figure 6).

miR-384 and COL10A1 expression in NSCLC tissues and adjacent tissues

Figure 6
miR-384 and COL10A1 expression in NSCLC tissues and adjacent tissues

(A) The relative mRNA expression of miR-384 and COL10A1 was detected using qRT-PCR. (B) Protein was isolated from NSCLC tissues and adjacent tissues and determined by Western blotting. (C) Quantitation of protein expression. Data are shown mean ± S.D. and experiments were repeated three times. @, P<0.05 when compared with adjacent tissues.

Figure 6
miR-384 and COL10A1 expression in NSCLC tissues and adjacent tissues

(A) The relative mRNA expression of miR-384 and COL10A1 was detected using qRT-PCR. (B) Protein was isolated from NSCLC tissues and adjacent tissues and determined by Western blotting. (C) Quantitation of protein expression. Data are shown mean ± S.D. and experiments were repeated three times. @, P<0.05 when compared with adjacent tissues.

The highest level of COL10A1 was found in A549 cells

qRT-PCR and Western blotting were applied to detect the COL10A1 expression in NSCLC cells and normal lung epithelial cells. Compared with normal cells, COL10A1 mRNA and proteins were highly expressed in A549, GLC82, MES-1, and TEP-s cell lines (all P<0.05). Compared with BEAS-2B, the levels of miR-384 in A549, MES-1, and TEP-s were significantly decreased (all P<0.05), which in GLC82 cells was no significant (P>0.05). And the highest level of COL10A1 and the lowest expression of miR-384 were found in A549 cells (all P<0.05). Thus, A549 cells were used for following experiments (Figure 7).

The expression of COL10A1 and miR-384 in NSCLC cells and pulmonary epithelial cells

Figure 7
The expression of COL10A1 and miR-384 in NSCLC cells and pulmonary epithelial cells

(A) The relative expression of miR-384 was detected by qRT-PCR. (B) The relative mRNA expression of COL10A1 was detected by qRT-PCR. (C,D) The relative protein expression of COL10A1 was detected by Western blotting. Data are shown as mean ± S.D. and experiments were repeated three times. #, P<0.05 when compared with Control group; $, P<0.05 when compared with A549 group.

Figure 7
The expression of COL10A1 and miR-384 in NSCLC cells and pulmonary epithelial cells

(A) The relative expression of miR-384 was detected by qRT-PCR. (B) The relative mRNA expression of COL10A1 was detected by qRT-PCR. (C,D) The relative protein expression of COL10A1 was detected by Western blotting. Data are shown as mean ± S.D. and experiments were repeated three times. #, P<0.05 when compared with Control group; $, P<0.05 when compared with A549 group.

siRNA-COL10A1-3 had the best interference capability

Pre-experiment for COL10A1 siRNA selection had been performed. Compared with siRNA NC, expression of COL10A1 for siRNA-COL10A1-2 and siRNA-COL10A1-3 was obviously down-regulated (all P<0.05). Compared with siRNA-COL10A1-2, siRNA-COL10A1-3 markedly down-regulated the level of COL10A1 protein (P<0.05). siRNA-COL10A1-3 was used for following experiments (Figure 8).

Level of COL10A1 protein interfered by different siRNAs

Figure 8
Level of COL10A1 protein interfered by different siRNAs

Data are shown as mean ± S.D. and experiments were repeated three times. , P<0.05 when compared with siRNA-COL10A1-NC. %, P<0.05 when compared with siRNA-COL10A1-2.

Figure 8
Level of COL10A1 protein interfered by different siRNAs

Data are shown as mean ± S.D. and experiments were repeated three times. , P<0.05 when compared with siRNA-COL10A1-NC. %, P<0.05 when compared with siRNA-COL10A1-2.

Overexpressed miR-384 or silenced COL10A1 suppressed proliferation of A549

To evaluate the impact of miR-384 and COL10A1 in NSCLC cell proliferation at 24, 48, and 72 h, MTT assay was carried out after cell transfection for 48 h. As shown in Figure 9, the NSCLC cell proliferation showed little difference in 24 h amongst all the groups. And no significant difference was observed amongst blank, NC, and miR-384 inhibitor+siRNA-COL10A1 groups at each time point (all P>0.05). Compared with blank group, a notably decreasing cell proliferation was detected in miR-384 mimic and siRNA-COL10A1 groups at 48 and 72 h, while that in miR-384 inhibitor group was improved (all P<0.05). These results indicated that miR-384 might be involved in the NSCLC cell proliferation by down-regulating COL10A1 and the effect of miR-384 inhibitor on A549 proliferation can be rescued by knockdown of COL10A1.

A549 cells vitality in each group

Figure 9
A549 cells vitality in each group

The cell activity was measured using MTT assay. Data are shown as mean ± S.D. and experiments were repeated three times, &, P<0.05 when compared with blank group.

Figure 9
A549 cells vitality in each group

The cell activity was measured using MTT assay. Data are shown as mean ± S.D. and experiments were repeated three times, &, P<0.05 when compared with blank group.

Overexpressed miR-384 or silenced COL10A1 could promote A549 apoptosis

After transfection for 48 h, qRT-PCR, Western blot, and Annexin-V-FITC/PI staining were employed to detect the cell apoptosis in each group. As identified in Figures 10 and 11, compared with blank group, no significant difference in expressions of miR-384, COL10A1, Survivin, Bcl-2, Bcl-xl and Bax, and apoptotic rate was observed in NC and miR-384 inhibitor+siRNA-COL10A1 groups (all P>0.05). But in miR-384 mimic and siRNA-COL10A1 groups, the miR-384, Bax mRNA and protein expressions were up-regulated while the expressions of COL10A1, Survivin, Bcl-2, and Bcl-xl were significantly reduced and the cell apoptosis rate was subsequently increased compared with blank group (all P<0.05). In contrast, evidently decreasing expressions of miR-384 and Bax and increasing mRNA and protein expressions of COL10A1, Survivin, Bcl-2, and Bcl-xl were observed and the cell apoptosis was induced after inhibition of miR-384 as well (all P<0.05). These results suggested that overexpressed miR-384 could suppress cell apoptosis by inhibiting COL10A1. The suppression of miR-384 inhibitor on A549 apoptosis can be neutralized by silence of COL10A1.

Relative expressions of Survivin, Bcl-2, Bax, and Bcl-xl of A549 cells in each group

Figure 10
Relative expressions of Survivin, Bcl-2, Bax, and Bcl-xl of A549 cells in each group

(A) Total RNA was collected after transfection for 48 h and mRNA expressions of Survivin, Bcl-2, Bax, and Bcl-xl in each group were measured using qRT-PCR. (B) Bands of above proteins. (C) Western blotting for protein expressions of NSCLC cells apoptosis-related factors including Survivin, Bcl-2, Bax, and Bcl-xl after transfection for 48 h. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Figure 10
Relative expressions of Survivin, Bcl-2, Bax, and Bcl-xl of A549 cells in each group

(A) Total RNA was collected after transfection for 48 h and mRNA expressions of Survivin, Bcl-2, Bax, and Bcl-xl in each group were measured using qRT-PCR. (B) Bands of above proteins. (C) Western blotting for protein expressions of NSCLC cells apoptosis-related factors including Survivin, Bcl-2, Bax, and Bcl-xl after transfection for 48 h. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

A549 cell apoptosis determined by Annexin V-FITC/PI staining

Figure 11
A549 cell apoptosis determined by Annexin V-FITC/PI staining

(A) Image output by flow cytometer; (B) cell apoptosis after transfection for 48 h was detected using Annexin V-FITC/PI double staining. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Figure 11
A549 cell apoptosis determined by Annexin V-FITC/PI staining

(A) Image output by flow cytometer; (B) cell apoptosis after transfection for 48 h was detected using Annexin V-FITC/PI double staining. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Overexpressed miR-384 or silenced COL10A1 could boost cell autophagy

To further detect cell autophagy in each group, qRT-PCR and Western blot were used to detect the biomarkers (Beclin1 and LC3B) in cell autophagy. As illustrated in Figure 12, the expression of Beclin1, LC3B mRNA, and LC3II/LC3I ratio showed no significant difference amongst blank, NC, and miR-384 inhibitor+siRNA-COL10A1 groups (all P>0.05). Compared with blank group, the Beclin1 and LC3B mRNA expression, Beclin 1 protein expression and LC3II/LC3I ratio were improved in miR-384 mimic and siRNA-COL10A1 groups (all P<0.05). But these indicators were all decreased in miR-384 inhibitor group (all P<0.05).

Expressions of Beclin1 and LC3 in A549 cells

Figure 12
Expressions of Beclin1 and LC3 in A549 cells

(A) Total RNA was extracted after transfection for 48 h. Beclin1 and LC3 mRNA expressions of A549 cell were detected using qRT-PCR. (B) Bands of above proteins. (C) Protein expressions of Beclin1, LC3II and LC3I in each group were detected by Western blotting. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Figure 12
Expressions of Beclin1 and LC3 in A549 cells

(A) Total RNA was extracted after transfection for 48 h. Beclin1 and LC3 mRNA expressions of A549 cell were detected using qRT-PCR. (B) Bands of above proteins. (C) Protein expressions of Beclin1, LC3II and LC3I in each group were detected by Western blotting. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Immunofluorescence was adopted to detect level of LC3II in each group, and results were shown in Figure 13. Compared with blank group, the amount of LC3II in NC and miR-384 inhibitor+siRNA-COL10A1 groups showed no obvious difference (all P>0.05). While the fluorescence intensity of LC3II was enhanced in miR-384 mimic and siRNA-COL10A1 groups while weakened in miR-384 inhibitor group (all P<0.05).

Expression of LC3II in A549 cells determined by immunofluorescence

Figure 13
Expression of LC3II in A549 cells determined by immunofluorescence

(A) Image of LC3II by immunofluorescence (×200); (B) fluorescence intensity was detected 48 h after transfection. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Figure 13
Expression of LC3II in A549 cells determined by immunofluorescence

(A) Image of LC3II by immunofluorescence (×200); (B) fluorescence intensity was detected 48 h after transfection. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Moreover, acidic autophagic lysosomes were observed by MDC staining and the results (Figure 14) proved that numbers of acidic autophagic lysosomes exhibited no significant difference amongst blank, NC and miR-384 inhibitor+siRNA-COL10A1 groups (all P>0.05). Compared with blank groups, formation of autophagic lysosomes was significantly induced in miR-384 mimic and siRNA-COL10A1 groups, while that in miR-384 inhibitor group was significantly inhibited (all P<0.05). The above results demonstrated that the overexpression of miR-384 or reduction in COL10A1 can promote cell autophagy. Repression of miR-384 is able to reduce A549 autophagy, which can be eliminated by deletion of COL10A1.

Autophagy in A549 cells was detected by MDC staining

Figure 14
Autophagy in A549 cells was detected by MDC staining

(A) Fluorescence image of cell autophagy (×400); (B) formation of autolysosome was detected using MDC staining after transfection for 48 h. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Figure 14
Autophagy in A549 cells was detected by MDC staining

(A) Fluorescence image of cell autophagy (×400); (B) formation of autolysosome was detected using MDC staining after transfection for 48 h. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Overexpressed miR-384 or silenced COL10A1 could suppress tumor growth

Subsequently, the tumorigenesis of NSCLC cells was observed and quantitated. The results revealed that there was no significant difference in volume of subcutaneous tumors amongst blank, NC and miR-384 inhibitor+siRNA-COL10A1 groups at each time point (all P>0.05). The subcutaneous tumors in miR-384 inhibitor group presented an obvious growth after 21 days in comparison with blank group, but growth of tumors in miR-384 mimic and siRNA-COL10A1 groups was suppressed (all P<0.05) (Figure 15). The aforementioned findings demonstrated that overexpressed miR-384 could inhibit tumor growth through the negative regulation of COL10A1. The influence on tumorigenesis of A549 after inhibition of miR-384 can be antagonized by down-regulation of COL10A1.

The volume of subcutaneous tumors in nude mice at different time points

Figure 15
The volume of subcutaneous tumors in nude mice at different time points

(A) The tumor volume of nude mice in each group was measured at 35th day; (B) fold line diagram shows that the tumor volumes of nude mice gradually increase with the extension of time. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Figure 15
The volume of subcutaneous tumors in nude mice at different time points

(A) The tumor volume of nude mice in each group was measured at 35th day; (B) fold line diagram shows that the tumor volumes of nude mice gradually increase with the extension of time. Data are shown as mean ± S.D. and experiments were repeated three times. &, P<0.05 when compared with blank group.

Discussion

NSCLC is one of the most common carcinoma and there is an ever-rising prevalence in recent years in the crowd [23]. Based on its features of long latency, low survival rate, and high morbidity and mortality, NSCLC has become one of the most harmful malignant tumor to humans [24]. Hence, it is of great significance to reveal the underlying pathological mechanisms of NSCLC, find novel diagnostic biomarkers, and generate therapeutic strategies. Accumulating evidence indicates that miRNA plays an important role in the occurrence and progression of NSCLC, but the specific pathogenesis of NSCLC is not explicit [25]. In the present study, analysis through bioinformatics methods was performed on predicting potential modulators for NSCLC, which revealed that compared with adjacent tissues, level of COL10A1 was extremely highly expressed in lung cancer tissues in GSE19804 and GSE27262 microarrays. Then three miRNA prediction tools were used to elucidate miRNA that targetted COL10A1, and miR-384 was ultimately chosen. Findings obtained from the bioinformatics analysis and luciferase reporter assay indicate that miR-384 might be involved in the progession of NSCLC by negatively regulating COL10A1 expression. Finally, qRT-PCR, Western blotting, MTT assay, Annexin-V-FITC/PI staining, immunofluorescence, MDC staining, and tumorigenicity assay were performed and the results indicated that miR-384 could promote the apoptosis and autophagy of NSCLC cells by inhibiting COL10A1.

Previous studies indicate that miR-384 is implicated in occurrence and development of cardiovascular and neurological diseases [11,26]. For the last few years, an increasing number of researches have suggested that miR-384 is also involved in the pathogenesis of many cancers and functions as a tumor suppressor [18,27]. Plieskatt et al. [28] have found that miR-384 was markedly down-regulated in liver cancer tissues. A study carried out by Wang et al. [29] has revealed that miR-384 is involved in the progression of colorectal cancer, which can inhibit tumor invasion and metastasis by interacting with KRAS and CDC42. Another study in vitro shows that miR-384 level is inversely correlated with the growth and invasion of NSCLC cells [12]. However, to a great extent, downstream molecular mechanism of miR-384 in the development of NSCLC still remains unknown. The identification of COL10A1 as miR-384 target gene in our research might be an important addition to understating the molecular mechanism of miR-384 in NSCLC.

COL10A1, which is associated with the growth of articular chondrocytes, has been reported to be involved in various kinds of bone diseases [30,31]. Accumulating evidence suggests that COL10A1 is likely to play a critical role in the progression of tumors [13,32]. The platelet-derived growth factor receptor (PDGFRL) was regarded as a tumor suppressing gene and Kawata et al. [33] have found that the expression of COL10A1 was significantly reduced in chondrocytes dealing with overexpressed PDGFRL. Kaczkowski et al. [34] collected peripheral blood from a variety of tumor patients including breast cancer, lung cancer, colorectal cancer, bladder cancer, and pancreatic cancer and discovered that COL10A1 expression in different kinds of cancer patients revealed an evident increase compared with normal controls. Brodsky et al. [35] confirmed that the up-regulation of COL10A1 is closely associated with poor prognosis of breast cancer when analysing the gene expression of patients treated with HER2-targeted therapy and neoadjuvant chemotherapy. These discoveries manifested that COL10A1 could serve as a cancer biomarker and play a vital impact in monitoring tumor progression and tumor treatment.

Quite a number of studies have identified that COL10A1 is regulated by multiple miRNAs in cancer pathogenesis [36,37]. The data presented in our study demonstrate that COL10A1 was a downstream target of miR-384. Furthermore, results acquired from experiments in vivo and in vitro notably indicate the NSCLC cell apoptosis and autophagy were enhanced in miR-384 mimic and siRNA-COL10A1 groups, but cell proliferation was significantly suppressed. At the same time, the tumorigenic ability in nude rats was significantly weaker than that in other groups. However, biological characteristics of NSCLC cells in blank, NC, and miR-384 inhibitor+siRNA-COL10A1 group exhibited no significant difference. These outcomes demonstrated that high expression of miR-384 contributes to NSCLC cell apoptosis and autophagy.

In summary, our data demonstrated that miR-384 is able to promote NSCLC cell apoptosis and autophagy through down-regulation of COL10A1. miR-384 is likely to be a potential therapeutic target against NSCLC as well as a biomarker associated with prognosis of NSCLC. Molecular mechanism of the present study has been displayed in Figure 16. Nevertheless, there might still be some potential factors such as lack of studies in vivo affecting our research, so more animal experiments about surpression of miR-384 in NSCLC cells should be carried out in the future to further identify our conclusion. Studies on the mechanism of miR-384 in NSCLC cells are not competed, and its downstream mechanisms, such as signaling pathways, need to be further studied.

Molecular mechanism of miR-384 down-regulating the level of COL10A1 to promote autophagy and apoptosis of NSCLC cells

Author contribution

Q.G. and M.Z. designed the study, collected, analyzed and interpreted the data, and wrote this article. Y.X., N.W., and W.Z. participated in analyzing and interpreting the data, and critically revised the article.

Funding

This work was supported by the Shanghai Municipal Science and Technology Commission Medical Guidance Project [grant number 124119b1900]; and the Shanghai Municipal Health and Family Planning Commission Program [grant number 201640197].

Ethics statement

All experiments were conducted under the guidelines and principles of the Declaration of Helsinki. The experimental program was approved by the Ethics committee of Tongren Hospital, Shanghai Jiao Tong University, School of Medicine and all study subjects had signed written informed consent before experiment. Meanwhile, all experimental operations followed the International Convention on Laboratory Animal Ethics and complied with relevant national regulations. All efforts were made to minimize animal suffering.

Competing interests

The authors declare that there are no competing interests associated with the manuscript.

Abbreviations

     
  • COL10A1

    collagen α-1(X) chain

  •  
  • GAPDH

    glyceraldehyde phosphate dehydrogenase

  •  
  • MDC

    monodansylcadaverine

  •  
  • NC

    negative control

  •  
  • NSCLC

    non-small cell lung cancer

  •  
  • PDGFRL

    platelet-derived growth factor receptor

  •  
  • PI

    propidium iodide

  •  
  • qRT-PCR

    quantitative real-time PCR

  •  
  • SPF

    specific-pathogen free

  •  
  • TNM

    tumor node metastasis

References

References
1.
Lin
T.C.
,
Lin
P.L.
,
Cheng
Y.W.
,
Wu
T.C.
,
Chou
M.C.
,
Chen
C.Y.
et al. .
(
2015
)
MicroRNA-184 deregulated by the microRNA-21 promotes tumor malignancy and poor outcomes in non-small cell lung cancer via targeting CDC25A and c-Myc
.
Ann. Surg. Oncol.
22
,
S1532
S1539
[PubMed]
2.
Kim
K.H.
,
Park
T.Y.
,
Lee
J.Y.
,
Lee
S.M.
,
Yim
J.J.
,
Yoo
C.G.
et al. .
(
2014
)
Prognostic significance of initial platelet counts and fibrinogen level in advanced non-small cell lung cancer
.
J. Korean Med. Sci.
29
,
507
511
[PubMed]
3.
Wolff
H.B.
,
Alberts
L.
,
Kastelijn
E.A.
,
Lissenberg-Witte
B.I.
,
Twisk
J.W.
,
Lagerwaard
F.J.
et al. .
(
2018
)
Differences in longitudinal health utility between stereotactic body radiation therapy and surgery in stage I non-small cell lung cancer
.
J. Thorac. Oncol.
13
,
689
698
[PubMed]
4.
Liu
D.
,
Nakashima
N.
,
Nakano
J.
,
Tarumi
S.
,
Matsuura
N.
,
Nakano
T.
et al. .
(
2017
)
Customized adjuvant chemotherapy based on biomarker examination may improve survival of patients completely resected for non-small-cell lung cancer
.
Anticancer Res.
37
,
2501
2507
[PubMed]
5.
Giaj-Levra
N.
,
Ricchetti
F.
and
Alongi
F.
(
2016
)
What is changing in radiotherapy for the treatment of locally advanced nonsmall cell lung cancer patients? A review
.
Cancer Invest.
34
,
80
93
[PubMed]
6.
MacDonagh
L.
,
Gray
S.G.
,
Finn
S.P.
,
Cuffe
S.
,
O’Byrne
K.J.
and
Barr
M.P.
(
2015
)
The emerging role of microRNAs in resistance to lung cancer treatments
.
Cancer Treat. Rev.
41
,
160
169
[PubMed]
7.
Kefas
B.
,
Comeau
L.
,
Floyd
D.H.
,
Seleverstov
O.
,
Godlewski
J.
,
Schmittgen
T.
et al. .
(
2009
)
The neuronal microRNA miR-326 acts in a feedback loop with notch and has therapeutic potential against brain tumors
.
J. Neurosci.
29
,
15161
15168
[PubMed]
8.
Gits
C.M.
,
van Kuijk
P.F.
,
Jonkers
M.B.
,
Boersma
A.W.
,
Smid
M.
,
van Ijcken
W.F.
et al. .
(
2014
)
MicroRNA expression profiles distinguish liposarcoma subtypes and implicate miR-145 and miR-451 as tumor suppressors
.
Int. J. Cancer
135
,
348
361
[PubMed]
9.
Braga
E.A.
,
Loginov
V.I.
,
Pronina
I.V.
,
Khodyrev
D.S.
,
Rykov
S.V.
,
Burdennyy
A.M.
et al. .
(
2015
)
Upregulation of RHOA and NKIRAS1 genes in lung tumors is associated with loss of their methylation as well as with methylation of regulatory miRNA genes
.
Biochemistry (Mosc.)
80
,
483
494
[PubMed]
10.
Wang
G.
,
Pan
J.
,
Zhang
L.
,
Wei
Y.
and
Wang
C.
(
2017
)
Long non-coding RNA CRNDE sponges miR-384 to promote proliferation and metastasis of pancreatic cancer cells through upregulating IRS1
.
Cell Prolif.
50
,
e12389
,
[PubMed]
11.
Song
H.
,
Rao
Y.
,
Zhang
G.
and
Kong
X.
(
2018
)
MicroRNA-384 inhibits the growth and invasion of renal cell carcinoma cells by targeting astrocyte elevated gene 1
.
Oncol. Res.
26
,
457
466
[PubMed]
12.
Fan
N.
,
Zhang
J.
,
Cheng
C.
,
Zhang
X.
,
Feng
J.
and
Kong
R.
(
2017
)
MicroRNA-384 represses the growth and invasion of non-small-cell lung cancer by targeting astrocyte elevated gene-1/Wnt signaling
.
Biomed. Pharmacother.
95
,
1331
1337
[PubMed]
13.
Chapman
K.B.
,
Prendes
M.J.
,
Sternberg
H.
,
Kidd
J.L.
,
Funk
W.D.
,
Wagner
J.
et al. .
(
2012
)
COL10A1 expression is elevated in diverse solid tumor types and is associated with tumor vasculature
.
Future Oncol.
8
,
1031
1040
[PubMed]
14.
Wu
X.
,
Zhang
W.
,
Hu
Y.
and
Yi
X.
(
2015
)
Bioinformatics approach reveals systematic mechanism underlying lung adenocarcinoma
.
Tumori
101
,
281
286
[PubMed]
15.
Li
T.
,
Huang
H.
,
Shi
G.
,
Zhao
L.
,
Zhang
Z.
,
Liu
R.
et al. .
(
2018
)
TGF-beta1-SOX9 axis-inducible COL10A1 promotes invasion and metastasis in gastric cancer via epithelial-to-mesenchymal transition
.
Cell Death Dis.
9
,
849
[PubMed]
16.
Huang
H.
,
Li
T.
,
Ye
G.
,
Zhao
L.
,
Zhang
Z.
,
Mo
D.
et al. .
(
2018
)
High expression of COL10A1 is associated with poor prognosis in colorectal cancer
.
Onco. Targets Ther.
11
,
1571
1581
[PubMed]
17.
Smyth
G.K.
(
2004
)
Linear models and empirical bayes methods for assessing differential expression in microarray experiments
.
Stat. Appl. Genet Mol. Biol.
3
,
3
[PubMed]
18.
Wang
Y.
,
Zhang
Z.
and
Wang
J.
(
2018
)
MicroRNA-384 inhibits the progression of breast cancer by targeting ACVR1
.
Oncol. Rep.
39
,
2563
2574
[PubMed]
19.
Wang
J.
and
Li
H.
(
2018
)
CircRNA circ_0067934 silencing inhibits the proliferation, migration and invasion of NSCLC cells and correlates with unfavorable prognosis in NSCLC
.
Eur. Rev. Med. Pharmacol. Sci.
22
,
3053
3060
[PubMed]
20.
Livak
K.J.
and
Schmittgen
T.D.
(
2001
)
Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method
.
Methods
25
,
402
408
[PubMed]
21.
Sun
H.
,
He
L.
,
Ma
L.
,
Lu
T.
,
Wei
J.
,
Xie
K.
et al. .
(
2017
)
LncRNA CRNDE promotes cell proliferation, invasion and migration by competitively binding miR-384 in papillary thyroid cancer
.
Oncotarget
8
,
110552
110565
[PubMed]
22.
Lai
Y.Y.
,
Shen
F.
,
Cai
W.S.
,
Chen
J.W.
,
Feng
J.H.
,
Cao
J.
et al. .
(
2016
)
MiR-384 regulated IRS1 expression and suppressed cell proliferation of human hepatocellular carcinoma
.
Tumour Biol.
37
,
14165
14171
[PubMed]
23.
Baik
C.S.
,
Chamberlain
M.C.
and
Chow
L.Q.
(
2015
)
Targeted therapy for brain metastases in EGFR-mutated and ALK-rearranged non-small-cell lung cancer
.
J. Thorac. Oncol.
10
,
1268
1278
[PubMed]
24.
Minguet
J.
,
Smith
K.H.
and
Bramlage
P.
(
2016
)
Targeted therapies for treatment of non-small cell lung cancer–Recent advances and future perspectives
.
Int. J. Cancer
138
,
2549
2561
[PubMed]
25.
Yin
Z.
,
Xu
M.
and
Li
P.
(
2017
)
miRNA-221 acts as an oncogenic role by directly targeting TIMP2 in non-small-cell lung carcinoma
.
Gene
620
,
46
53
[PubMed]
26.
Gu
Q.H.
,
Yu
D.
,
Hu
Z.
,
Liu
X.
,
Yang
Y.
,
Luo
Y.
et al. .
(
2015
)
miR-26a and miR-384-5p are required for LTP maintenance and spine enlargement
.
Nat. Commun.
6
,
6789
[PubMed]
27.
Zhang
J.F.
,
Zhang
G.Y.
,
Hu
X.M.
,
Luo
Z.P.
and
Ma
Y.Z.
(
2018
)
MicroRNA-384 downregulates SETD8 expression to suppress cell growth and metastasis in osteosarcoma cells
.
Eur. Rev. Med. Pharmacol. Sci.
22
,
1602
1608
[PubMed]
28.
Plieskatt
J.L.
,
Rinaldi
G.
,
Feng
Y.
,
Peng
J.
,
Yonglitthipagon
P.
,
Easley
S.
et al. .
(
2014
)
Distinct miRNA signatures associate with subtypes of cholangiocarcinoma from infection with the tumorigenic liver fluke Opisthorchis viverrini
.
J. Hepatol.
61
,
850
858
[PubMed]
29.
Wang
Y.X.
,
Chen
Y.R.
,
Liu
S.S.
,
Ye
Y.P.
,
Jiao
H.L.
,
Wang
S.Y.
et al. .
(
2016
)
MiR-384 inhibits human colorectal cancer metastasis by targeting KRAS and CDC42
.
Oncotarget
7
,
84826
84838
[PubMed]
30.
Dehne
T.
,
Schenk
R.
,
Perka
C.
,
Morawietz
L.
,
Pruss
A.
,
Sittinger
M.
et al. .
(
2010
)
Gene expression profiling of primary human articular chondrocytes in high-density micromasses reveals patterns of recovery, maintenance, re- and dedifferentiation
.
Gene
462
,
8
17
[PubMed]
31.
Ray
S.
,
Thormann
U.
,
Sommer
U.
,
Khassawna
T.E.
,
Hundgeburth
M.
,
Henss
A.
et al. .
(
2016
)
Effects of macroporous, strontium loaded xerogel-scaffolds on new bone formation in critical-size metaphyseal fracture defects in ovariectomized rats
.
Injury
47
(
Suppl. 1
),
S52
S61
[PubMed]
32.
Sole
X.
,
Crous-Bou
M.
,
Cordero
D.
,
Olivares
D.
,
Guino
E.
,
Sanz-Pamplona
R.
et al. .
(
2014
)
Discovery and validation of new potential biomarkers for early detection of colon cancer
.
PLoS ONE
9
,
e106748
[PubMed]
33.
Kawata
K.
,
Kubota
S.
,
Eguchi
T.
,
Aoyama
E.
,
Moritani
N.H.
,
Oka
M.
et al. .
(
2017
)
A tumor suppressor gene product, platelet-derived growth factor receptor-like protein controls chondrocyte proliferation and differentiation
.
J. Cell. Biochem.
118
,
4033
4044
[PubMed]
34.
Kaczkowski
B.
,
Tanaka
Y.
,
Kawaji
H.
,
Sandelin
A.
,
Andersson
R.
,
Itoh
M.
et al. .
(
2016
)
Transcriptome analysis of recurrently deregulated genes across multiple cancers identifies new pan-cancer biomarkers
.
Cancer Res.
76
,
216
226
[PubMed]
35.
Brodsky
A.S.
,
Xiong
J.
,
Yang
D.
,
Schorl
C.
,
Fenton
M.A.
,
Graves
T.A.
et al. .
(
2016
)
Identification of stromal ColXalpha1 and tumor-infiltrating lymphocytes as putative predictive markers of neoadjuvant therapy in estrogen receptor-positive/HER2-positive breast cancer
.
BMC Cancer
16
,
274
[PubMed]
36.
Weber
M.
,
Sotoca
A.M.
,
Kupfer
P.
,
Guthke
R.
and
van Zoelen
E.J.
(
2013
)
Dynamic modelling of microRNA regulation during mesenchymal stem cell differentiation
.
BMC Syst. Biol.
7
,
124
[PubMed]
37.
Yang
Z.
,
Hao
J.
and
Hu
Z.M.
(
2015
)
MicroRNA expression profiles in human adipose-derived stem cells during chondrogenic differentiation
.
Int. J. Mol. Med.
35
,
579
586
[PubMed]
This is an open access article published by Portland Press Limited on behalf of the Biochemical Society and distributed under the Creative Commons Attribution License 4.0 (CC BY).