Background
MicroRNAs (miRNA) are a class of small noncoding RNAs of 18–24 nucleotides that bind to partially complementary recognition sequences of mRNA, causing either degradation or inhibition of translation and thus effectively silencing their mRNA target [
1]. To date, more than 1881 human miRNAs have been registered in miRBase (Release 21;
http://www.mirbase.org/). Massively parallel signature sequencing of miRNAs can identify the miRNome in-depth, revealing differences in miRNA expression as well as individual miRNA abundances. Recent evidence suggests that dysregulation of miRNAs is involved in several biological processes, such as development, differentiation, proliferation, and apoptosis [
2,
3]. miRNAs have also been suggested to play essential roles in initiation and progression of certain types of cancer, such as lung cancer [
4,
5], breast cancer [
6], and colorectal cancer [
7]. Although classical approaches to cloning have partially revealed miRNA expression profiles in a panel of mammalian tissues and cell types [
8,
9], low throughput, low sensitivity, and poor resolution make these approaches unlikely to define the miRNome.
Lung cancer is one of the most frequently diagnosed cancers and the leading cause of cancer-associated death both in men and women worldwide [
10]. It was estimated that there were 1.8 million new cases in 2012, leading to the death of approximately 1.59 million people per year globally, extrapolating from a 2012 International Agency for Research on Cancer risk assessment [
11], and this trend is expected to continue until 2030. Approximately 85% of lung cancers are classified histopathologically as non-small cell lung carcinomas (NSCLCs) [
12]. Although treatment advances have been achieved with the use of surgery and chemotherapy [
13], and targeted therapies (EGFR or ALK tyrosine kinase inhibitors) have significantly improved the outcomes of some subgroups of advanced NSCLC patients, the 5-year overall survival (OS) rate is only 16% for all stages of the disease [
14]. The poor outcomes and frequent relapses associated with lung cancer urgently demand the development of new screening methods and early biomarkers for accurate and non-invasive detection of lung cancer metastasis and recurrence [
15,
16]. Therefore, there is a great need for the development of strategies for improved diagnosis, prevention, and therapy for NSCLC.
Several studies have shown that miRNAs can be used as diagnostic and prognostic biomarkers [
4‐
7,
17‐
23]. However, the majority of previous studies have been based on small sample sizes, relatively limited numbers of miRNAs, or limited miRNA detection methods [
5,
19‐
23]. In NSCLC, several miRNAs are essential for tumor development, including miR-9, let-7, and miR-193a-3p [
24‐
26]. Downregulation of miRNAs, such as miR-10b, − 143, −181a, − 21, − 205, − 221, − 222, and -449a, has been shown to be a key factor in NSCLC tumorigenesis [
27]. However, the expressions of other miRNAs, including miR-106a, − 146, − 155, − 150, − 17-3p, − 191, − 197, − 192, − 21, − 203, − 205, − 210, − 212, and − 214, have been reported to be upregulated in lung cancer [
5]. Our previous study of 32 lung adenocarcinoma patients with brain metastasis (BM) versus 55 patients without BM using miRNA array profiles demonstrated that miR-4317 expression levels can predict OS in NSCLC. However, the precise molecular mechanism through which miR-4317 influences NSCLC progression remains largely unknown. In this study, we identified a new miRNA and signaling pathway for understanding the pathogenesis of NSCLC, providing a promising biomarker and therapeutic target for NSCLC.
Methods
Human NSCLC tissue samples
A chip including 162 cases of non-metastatic NSCLC tissues and non-neoplastic lung tissues was purchased from Outdo Biotech (HlugA180Su02, Shanghai, China; http://
www.superchip.com.cn/). Another 140 paired NSCLC tissues and matched adjacent non-cancerous tissues were collected in North China University of Science and Technology Affiliated People’s Hospital from 2012 to 2015. Histologic characteristics of these samples are summarized in Table
1. Serum samples from 86 NSCLC patients and 40 healthy controls were obtained from the abovementioned hospital (Table
2). All serum specimens were transported at 4 °C and stored at − 80 °C until RNA extraction. None of the patients had received preoperative treatments such as radiotherapy or chemotherapy. The methods used to treat the tissues were carried out strictly in accordance with institutional policies and approved guidelines for experiment procedures.
Table 1
Correlations between miR-4317 expression and clinicopathological features in patients with NSCLC
Gender |
Male | 89 (74.2) | 28 (66.7) | 0.35 | 65 (63.7) | 26 (68.4) | 0.60 |
Female | 31 (25.8) | 14 (23.3) | 37 (36.3) | 12 (31.6) |
Agea |
< 60 y | 45 (37.8) | 15 (35.7) | 0.81 | 43 (42.2) | 16 (42.1) | 1.00 |
≥ 60 y | 74 (62.2) | 27 (64.3) | 59 (57.8) | 22 (57.9) |
Tumor sizea |
< 5 cm | 74 (31.7) | 29 (48.5) | 0.39 | 69 (69.0) | 28 (73.7) | 0.59 |
≥ 5 cm | 46 (68.3) | 13 (51.5) | 31 (31.0) | 10 (26.3) |
Tumor stage |
T1 + T2 | 90 (75.0) | 35 (83.3) | 0.27 | 79 (77.5) | 33 (86.8) | 0.22 |
T3 + T4 | 30 (25.0) | 7 (16.7) | 23 (22.5) | 5 (13.2) |
Histological type |
Adenocarcinoma | 63 (52.5) | 24 (57.1) | 0.60 | 59 (57.8) | 17 (44.7) | 0.17 |
Squamous cell carcinoma | 57 (47.5) | 18 (41.9) | 43 (42.2) | 21 (55.3) |
Histological gradea |
Well/moderate | 88 (73.3) | 27 (64.3) | 0.27 | 57 (61.3) | 30 (78.9) | 0.05 |
Poor/NS | 32 (26.7) | 15 (35.7) | 36 (38.7) | 8 (21.1) |
Lymph node metastasisa |
Negative | 49 (51.6) | 32 (80.0) | 0.002 | 58 (59.8) | 33 (86.8) | 0.003 |
Positive | 46 (48.4) | 8 (20.0) | 39 (40.2) | 5 (18.5) |
TNM stagea |
I + II | 58 (49.6) | 36 (87.8) | 0.000 | 61 (60.4) | 34 (89.5) | 0.001 |
III + IV | 59 (50.4) | 5 (12.2) | 40 (39.6) | 4 (10.5) |
Table 2
Correlations between serum expression of miR-4317 and clinicopathological features in 86 patients with NSCLC
Gender |
Male | 27 (62.8) | 29 (67.4) | 0.65 |
Female | 16 (37.2) | 14 (22.6) |
Age |
< 60 y | 17 (65.4) | 19 (44.2) | 0.66 |
≥ 60 y | 26 (34.6) | 24 (55.8) |
Tumor size |
< 5 cm | 30 (69.8) | 33 (76.7) | 0.47 |
≥ 5 cm | 13 (30.2) | 10 (23.3) |
Tumor stage |
T1 + T2 | 31 (72.1) | 38 (88.4) | 0.06 |
T3 + T4 | 12 (27.9) | 5 (11.6) |
Histological type |
Adenocarcinoma | 30 (69.8) | 23 (53.5) | 0.12 |
Squamous cell carcinoma | 13 (30.2) | 20 (46.5) |
Histological gradea |
Well/moderate | 30 (71.4) | 28 (65.1) | 0.53 |
Poor/NS | 12 (28.6) | 15 (34.9) |
Lymph node metastasis |
Negative | 18 (41.9) | 34 (79.1) | < 0.0001 |
Positive | 25 (58.1) | 9 (21.9) |
Clinical stages |
I + II | 20 (46.5) | 33 (76.7) | 0.004 |
III + IV | 23 (54.5) | 10 (23.3) |
MicroRNA microarray assay
The analysis of miRNA microarray data was conducted in 32 clinical samples acquired from lung adenocarcinoma patients with BM in contrast to 55 patients without BM obtained from the Cancer Hospital, Chinese Academy of Medical Sciences (Beijing, China) between 2003 and 2008. In short, total RNA isolated from patient samples was examined with the mammalian miRNA array V2.0 (CapitalBio, Beijing, China), which identifies 1105 miRNAs in humans, mice, and rats. Separation of low-molecular-weight RNAs from total RNA was carried out by a PEG precipitation method; the low-molecular-weight RNAs were then labeled with 5-phosphate-cytidyl-uridyl-Cy3–3 and then hybridized to the mammalian miRNA array overnight at 42 °C. A LuxScan 10 K/A laser confocal scanner was used to scan the arrays, and the acquired images were evaluated using LuxScan 3.0 software (both from CapitalBio). Cluster 3.0 was used to carry out clustering analysis, and the results were viewed with TreeView software. The normalization of fluorescence signals was carried out using the median center tool for genes in Cluster 3.0, and they were evaluated using the significance analysis of microarrays (SAM), with a false discovery rate (FDR) threshold set of 0 and fold-change established at ≥2- or ≤ 0.5-fold change and p value < 0.05.
Cell lines and cell culture
All NSCLC cell lines used in this study, including A549, NCI-H1299, NCI-H157, ANIP-973, GLC-82, and NCI-H292, were cultured in 1640 RPMI medium supplemented with 10% fetal bovine serum at 37 °C in a humidified atmosphere containing 5% CO2. The human fetal lung fibroblast cell line (MRC-5) was cultured in Minimum Essential Medium (MEM) containing non-essential amino acids, Earle’s salts, and L-glutamine supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic solution (containing 100 U/mL penicillin, 100 μg/mL streptomycin, and 0.25 μg amphotericin), and was maintained in a humidified air atmosphere with 5% CO2 at 37 °C.
In situ hybridization (ISH) of miR-4317
ISH was performed per the manufacturer’s instructions. The miR-4317 probe was tagged with 3′ and 5′ digoxigenin and LNA modified (
Redlandbio.biomart.cn, Guangzhou, China). The probe-target complex was detected using an antidigoxigenin-alkaline phosphate conjugate and nitro-blue tetrazolium and 5-bromo-4-chloro-3′-indolyphosphate as the chromogen. Cases were classified according to the cytoplasmic miR-4317 intensity as follows: negative = negative or faint expression in most cells; low expression = low expression in most cells or moderate expression in < 50% of the cells; high expression = moderate to strong expression in most cells.
miRNA transfection
All endogenous mature miRNA mimics, inhibitors, and agomirs were purchased from RiboBio (Guangzhou, China). For transfection, experimental protocols were performed according to the manufacturer’s instructions (RiboBio). The miRNA mimics, miRNA inhibitors, and miRNA NC were transfected into cells using Lipofectamine 2000 (Invitrogen, Carlsbad, USA) according to the manufacturer’s instructions. After a 48-h transfection, the cells were used for further experiments.
Plasmid construction
pDonR223-FGF9, pDonR223-CCND2, and pDonR223-TGFBR1 plasmids carrying human FGF9, CCND2, and TGFBR1 genes were purchased from Changsha Axybio Bio-Tech Co., Ltd. (Changsha, China). The complete coding sequences of human FGF9, CCND2 and TGFBR1 were amplified from the pDonR223-FGF9, pDonR223-CCND2, and pDonR223-CCND2 plasmids, respectively. FGF9, CCND2, and TGFBR1 products and pEGFP-N1 plasmid were digested with XhoI and HindIII, and the fragments were purified and ligated with T4 DNA ligase. The ligated product was transformed into Top10 competent cells, and the positive clones were named pEGFP-N1- FGF9, pEGFP-N1- CCND2, and pEGFP-N1- TGFBR1.
Quantitative real-time polymerase chain reaction
To evaluate the expressions of miR-4317, FGF9, CCND2, and TGFBR1, respectively, total RNA was used for qRT-PCR on a Step One Plus real-time system (AB Applied Biosystems, Carlsbad, CA, USA). U6 and GAPDH were used as internal controls. All primers used in this study are listed in Additional file
1: Table S1.
Target prediction and luciferase reporter assays
Bioinformatics analysis was performed using miRDB (
http://www.mirdb.org/), miRanda (
http://www.microrna.org), and TargetScan (
http://www.targetscan.org/). The 3’-UTRs of human FGF9, CCND2, and TGFBR1 were amplified from human genomic DNA and individually inserted into pmiR-RB-REPORT™ (Ribobio) via the XhoI and NotI sites. Similarly, the FGF9, CCND2, and TGFBR1 3’-UTR mutant fragments were inserted into the pmiR-RB-REPORT™ control vector at the same sites. For the reporter assays, NSCLC cells were co-transfected with wild-type reporter plasmid and miR-4317 mimics. Firefly and
Renilla luciferase activities were measured in cell lysates using the dual-luciferase reporter assay system. Luciferase activity was measured 48 h post-transfection using the Dual-Glo Luciferase reporter system according to the manufacturer’s instructions (Promega, Madison, WI, USA). Firefly luciferase units were normalized against
Renilla luciferase units to control for the transfection efficiency.
In vitro cell proliferation assays
For the cell proliferation assays, cells were seeded into each well of a 96-well plate (5 × 103 per well), and cell proliferation was determined by MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3–carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium) according to the manufacturer’s instructions(Best Bio, China). MTS solution was added (20 μL/well) to each well and incubated at 37 °C for 2 h. The optical density of each sample was immediately measured using a microplate reader (Bio-Rad, Hercules, CA, USA) at 570 nm.
Cells were transfected with miR-4317 mimic or miR mimic NC, miR-4317 inhibitor or miR inhibitor NC, as described above. Twenty-four hours later, the transfected cells were trypsinized, counted, and replated at a density of 1 × 103 cells/10-cm dish. Ten days later, colonies resulting from the surviving cells were fixed with 3.7% methanol, stained with 0.1% crystal violet, and counted. Colonies containing at least 50 cells were scored. Each assay was performed in triplicate.
Transwell migration/invasion assay
In vitro cell migration assays were performed according to the manufacturer’s instructions using transwell chambers (8 μM pore size; Costar, New York, USA). Cells were allowed to grow to subconfluency (~ 75–80%) and were serum-starved for 24 h. After detachment with trypsin, the cells were washed with PBS and resuspended in serum-free medium. Next, 100 μL of the cell suspension (5 × 10 4 cells/mL) was added to the upper chamber. Complete medium was added to the bottom wells of the chambers. For screening, the cells that had not migrated after 24 h were removed from the upper face of the filters using cotton swabs, but the cells that had migrated were fixed with 5% glutaraldehyde solution to determine the number of migratory cells. The lower surfaces of the filters were stained with 0.25% Trypan Blue. Images of six different × 10 fields were captured from each membrane, and the number of migratory cells was counted. The mean of triplicate assays for each experimental condition was used. Similar inserts coated with Matrigel were used to evaluate the cell invasive potential in the invasion assay.
Flow cytometry analysis
FACS analysis was performed 48 h post-transfection. The cells were harvested, washed with cold PBS, fixed in 70% ethanol at − 20 °C for 24 h, stained with 50 μg/mL propidium iodide (4ABio, China), and analyzed using a FACS Calibur flow cytometer (BD Biosciences, MA, USA). The results were analyzed using ModFit software (BD Biosciences). Three independent assays were conducted.
Western blot analysis
For the western blot analyses, RIPA buffer containing protease inhibitors and phospha tase inhibitors (Roche, Basel, Switzerland) was used to prepare whole-cell lysates. Briefly, equal amounts of lysate were separated by SDS-polyacrylamide gel electrophoresis and then transferred to PVDF membranes (Millipore,Massachusetts, USA). After blocking with 5% bovine serum albumin, the membranes were probed with anti-FGF9 or CCND2 and anti-GAPDH (ab71395, ab226972, ab9485, Abcam, Cambridge, UK), followed by incubation with a horseradish peroxidase–conjugated secondary antibody [goat-anti-mouse IgG (1:2000) and goat-anti-rabbit IgG (1:3000)]. The proteins were visualized using Image Reader LAS-4000 (Fujifilm) and analyzed with Multi Gauge V3.2 software (GE Healthcare Life Sciences, USA).
Generation of stable cell lines
Recombinant lentiviral vectors containing the miRNA-4317 knockdown and irrelevant sequence were purchased from Xibeihongcheng Biotechnology (Beijing, China). Added to the lentivirus expression vectors was a luciferase and puromycin reporter gene driven by the EF1α promoter to indicate the infection efficiency in a timely manner. To construct lentiviral vectors, the precursor sequences for the miRNA-4317 and irrelevant sequence (negative control) were inserted into pHBLV-U6-MCS-EF1α-Luc-T2A-puromycin lentiviral vectors. The recombinant lentiviruses were packaged by co-transfection of HEK-293 T cells with pSPAX2 and pMD2.G with LipoFiter™ reagent. The supernatant with lentivirus particles was harvested at 48 and 72 h after transfection and filtered through 0.45-μm cellulose acetate filters (Millipore). Recombinant lentiviruses were concentrated by ultracentrifugation. To establish stable cell lines, NSCLC cells were transduced with lentivirus at an MOI of approximately 5 in the presence of 5 μg/ml polybrene. The supernatant was removed after 24 h and replaced with fresh complete culture medium. The infection efficiency was confirmed by RT-PCR 96 h after infection, and the cells were selected with 2 μg/ml puromycin for 2 weeks.
All animals received humane care in compliance with the “Guide for the Care and Use of Laboratory Animals” prepared by the Institute of Laboratory Animal Resources published by the National Institutes of Health and according to the Animal Experiment Guidelines of Samsung Biomedical Research Institute. The effect of miR-4317 on the tumorigenic and metastatic potential of NSCLC cells was analyzed in subcutaneous and systemic metastasis in vivo models via right subcutaneous tissue and tail vein injection, respectively. For the subcutaneous model, 4–6-week-old BALB/c nude mice received a subcutaneous injection of 1 × 106 transfected cells in the right hip. For the experimental metastasis in vivo model, transfected cancer cells (1 × 106 in 100 μL of HBSS) were injected directly into the tail vein. Six weeks later, the tumor colony in the subcutaneous tissue was observed by HE staining and histological examination. Bioluminescence images were collected to assess the growth and metastasis of the implanted tumor cells. To quantify the in vivo bioluminescence signal, the mice were anesthetized with isoflurane before in vivo imaging, and D-luciferin solution (in vivo imaging solutions, PerkinElmer, 150 mg/kg in PBS, Waltham, Massachusetts, USA) was injected intravenously for both subcutaneous and systemic xenografts. Bioluminescence images were acquired with the IVIS Spectrum imaging system (PerkinElmer,) 2–5 min after injection, and the captured images were quantified using the Living Image Software package (PerkinElmer/Caliper Life Sciences) by measuring the photon flux (photons/s/cm2/steradian) within a region of interest drawn around the bioluminescence signal.
Agomir treatment
The agomir and the miRNA negative control were synthesized by the Bioribo Company and applied as per the manufacturer’s instructions. A 10-nmol miR-4317 agomir and the miRNA negative control in 0.1 ml saline buffer were locally injected into the NSCLC cell-forming tumor mass once every 5 d for 6 weeks. After the treatment, the NSCLC cell-forming tumors were applied for the immunohistochemistry assay. The tumor size was monitored by measuring the length (L) and width (W) with calipers every 5 d, and the volumes were calculated using the following formula: (L × W2)/2. Mice were euthanized by cervical dislocation on day 42, and the tumors were excised and snap-frozen for protein and RNA extraction.
Evaluation of immunohistochemical staining
Sections were deparaffinized and boiled in 10 mM citrate buffer (pH 6.0) for antigen retrieval. Endogenous peroxidase was blocked with 3% H2O2. The slides were incubated with primary antibodies, including Ki-67 (1:50, ab156956, Abcam), FGF9 (1:50), and CCND2 (1:50), respectively, followed by HRP-labeled secondary antibody. They were then visualized with diaminobenzidine. The expression levels of FGF9 and CCND2 in cytoplasm were determined by the product score of the average percentage and intensity of positive cells under 5 random high-power fields. Scores for percentage: < 5% (0), 5–25% (1), 25–50% (2), 50–75% (3), and > 75% (4); scores for intensity: no staining (0), light brown (1), brown (2), and dark brown (3). For FGF9 and CCND2, scores of 0 and ≥ 1 were defined as negative and positive, respectively.
Statistical analysis
All measurement data are expressed as means ± standard deviation. Error bars represent the standard errors of the means. Student’s t test, χ2 test and repeated measures ANOVA were used to perform statistical comparisons. Log-rank test was used to analyze the effect of clinical variables and miRNAs on OS of patients. Multivariate Cox regression models were used to assess factors associated with OS in NSCLC. Receiver operating characteristic curves and the area under the curve were used to assess the feasibility of using serum miRNA as a diagnostic tool for detecting NSCLC. p < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS 16.0 software (SPSS Inc., USA).
Discussion
In this study, we evaluated the role of miR-4317 in lung carcinoma, as well as the underlying molecular mechanisms. First, we found that as a novel antitumor miRNA in NSCLC, miR-4317 was significantly downregulated in lung cancer tissues and exhibited expression levels that were inversely correlated with the clinical tumor stage and lymph node metastasis by ISH and qRT-PCR. Recently, miRNAs linked to the clinical tumor stage and lymph node metastasis of NSCLC have been identified, such as miR-10, miR-195, miR-338-3p, and miR-708-5p [
18,
28‐
30]. Endogenous circulating miRNAs have attracted significant attention regarding the diagnosis, prognosis, and metastasis of cancer. Importantly, in the present study, the serum miR-4317 level was demonstrated to be useful in delineating lung cancer stages due to the decreasing expression of miR-4317 in higher-stage cancers. Lymph nodes that contained metastatic tumors also showed downregulation of miR-4317 expression. To the best of our knowledge, this is the first report on the diagnostic and prognostic value of serum miR-4317. Diagnostic tests that involve noninvasive procedures, such as a simple blood draws, are highly desirable and patient friendly. A decreased level of serum miR-4317 could be used to detect lung cancer early and accurately to determine lung cancer aggressiveness. This serum-based approach is more advantageous than biopsy, an invasive procedure that serves as the main tool for lung cancer risk assessment. miRNAs have also been extensively investigated as prognostic factors [
31]. Our results showed that the high expression of miR-4317 might be closely associated with improved OS in NSCLC. Intriguingly, Zhou et al. demonstrated that a high expression level of miR-574-5p in serum was an independent poor prognostic risk factor in patients with SCLC [
32]. Therefore, our findings demonstrate that miR-4317 may play a tumor-inhibiting role in NSCLC and prompted us to investigate its exact functions.
To better understand the underlying role of miR-4317 in lung cancer, we explored the biological function of miR-4317 in vitro and in vivo. In the present study, we showed that overexpression of miR-4317 inhibited the proliferation, colony formation, migration, and invasion capacity of NSCLC cells. In contrast, downregulation of miR-4317 had the opposite effects. Moreover, dysregulation of miR-4317 induced tumorigenesis and lung and brain metastases in nude mice. These results indicate that miR-4317 plays a crucial role in the growth and metastasis of NSCLC. Metastasis-related death accounts for approximately 90% of cancer mortality [
33]. Accumulating evidence shows that miRNAs participate in tumor growth and metastatic processes, and a growing number of miRNAs have been found to be involved in lung cancer metastasis [
34,
35]. Joshi et al. demonstrated that miR-148a might act as a tumor suppressor, and it inhibits migration and invasion of the A549 NSCLC cell line [
36]. Upregulation of miR-150 results in a significant increase in tumor cell metastasis in vitro and lung metastases in a nude mouse xenograft model [
37]. Similarly, the inhibitory roles of various miRNAs in tumor growth and metastasis have been demonstrated in previous NSCLC studies [
30,
38,
39]. Furthermore, upregulation of miR-26b leads to inhibition of gastric cancer cell migration and invasion in vitro and lung metastasis formation in vivo [
40,
41].
The mechanism by which miRNAs alter gene expression remains controversial, but most studies suggest that miRNAs are primarily processed by the RNA-mediated interference machinery to trigger partial or complete target gene mRNA degradation [
42]. Our bioinformatics analysis revealed that miR-4317 could bind to the 3’UTRs of FGF9 and CCND2, and we observed that the expressions of FGF9 and CCND2 were repressed by miR-4317. FGF9, which is also known as glial activating factor, is one of 23 members of the highly conserved FGF family. As a secreted, glycosylated 26-kDa protein, it has mitogenic effects on a variety of cell types [
43]. FGF9 is also required during lung development for mesenchymal growth and epithelial branching, and inactivation of FGF9 in mice results in perinatal death due to respiratory insufficiency [
44‐
46]. FGF9 has been shown to be implicated in cancers, such as ovarian endometrioid adenocarcinoma [
43], hepatocellular carcinoma [
47], and prostate carcinoma [
48]. Hendrix et al. showed that FGF9 possesses oncogenic activity [
43]. Abdel-Rahman et al. confirmed that FGF9 activates a major intracellular effector of ERK MAP kinase [
49]. Deng et al. identified a novel mechanism of miR-26a in the suppression of gastric cancer growth and metastasis by direct targeting of FGF9 protein expression [
50]. miR-182 downregulation induces SMC differentiation, proliferation, and migration, and these processes are prevented by FGF9 knockdown; moreover, inhibition of FGF9 itself is able to suppress the dedifferentiation, proliferation, and migration of rat SMCs [
51]. Importantly, Suzuki et al. examined FGF9 concentrations in lung cancer patient serum by ELISA assay and found that the mean concentration was less than the detectable range in 15 patients with lung cancer and 8 with other lung diseases [
52]. Although FGF9 serum concentrations may not directly reflect the FGF9 concentration in specific patients with lung cancer, combining the above in vitro findings, we speculate that, if finding elevated levels of FGF9 in the blood of patients or mice with low miR-4317 would give further support to the conclusions of this article.
In this study, another new direct and functional target of miR-4317, CCND2, was also identified. CCND2 is a well-known cyclin that functions in the cell cycle, specifically in G1/S transition. We further conducted fluorescence activated cell sorting (FACS) analysis to confirm the role of miR-4317 as a negative regulator of the cell cycle and found that increased expression of this miRNA resulted in significant G0/G1 arrest and S phase reduction. Recent reporter assays have shown that CCND2 is targeted by let-7a, and that this interaction inhibits proliferation in human prostate cancer cells both in vitro and in vivo [
53]. Moreover, overexpression of CCND2 results in excessive proliferation in many cell types and enhances OSCC cell invasive ability in vitro and in vivo [
54]. In this study, we confirmed that FGF9 and CCND2 were direct targets of miR-4317 in NSCLC cells. To determine whether miR-4317 could suppress NSCLC growth and metastasis by repressing FGF9 and CCND2 expression, we found that FGF9 and CCND2 overexpression could rescue the invasion and growth defects of miR-4317. These results suggested that miR-4317 inhibited NSCLC growth and metastasis partly by targeting FGF9 and CCND2. However, some limitations remain in our study, in that FGF9 and CCND2 might be not the only two proteins influenced by miR-4317. Moreover, other miRNAs may regulate the expression of FGF9 and CCND2 in addition to miR-4317 in NSCLC. Therefore, our future work will be dedicated to examining these issues, specifically by knocking-down FGF9 or CCND2 to abolish the effects of miR-4317 transfection and investigating the relationship between other signaling pathways and FGF9 or CCND2 protein.
Additionally, the proliferative activities of the tumor cells were assessed via immunohistochemical staining for Ki-67. The Ki-67 staining intensities were decreased in tumors from the miR-4317-downregulated group. The results further revealed reduced FGF9 and CCND2 staining intensities in xenograft tumors in the miR-4317 agomir group compared with the NC group, indicating a decrease in tumor cell proliferation in response to the downregulation of miR-4317. Finally, we found that high expression levels of FGF9 and CCND2 in NSCLC tissues were inversely associated with miR-4317. However, there have been no reports concerning the relationship between FGF9 and CCND2. Intriguingly, a recent study suggested that miR-4317 could repress the proliferation of gastric cancer cells by targeting and suppressing ZNF322 [
55].