Long non-coding RNA SNHG15 promotes CDK14 expression via miR-486 to accelerate non-small cell lung cancer cells progression and metastasis.

Long non-coding RNAs (lncRNAs) have been validated to play important role in multiple cancers, including non-small cell lung cancer (NSCLC). In present study, our team investigate the biologic role of SNHG15 in the NSCLC tumorigenesis. LncRNA SNHG15 was significantly upregulated in NSCLC tissue samples and cells, and its overexpression was associated with poor prognosis of NSCLC patients. In vitro, loss-of-functional cellular experiments showed that SNHG15 silencing significantly inhibited the proliferation, promoted the apoptosis, and induced the cycle arrest at G0//G1 phase. In vivo, xenograft assay showed that SNHG15 silencing suppressed tumor growth of NSCLC cells. Besides, SNHG15 silencing decreased CDK14 protein expression both in vivo and vitro. Bioinformatics tools and luciferase reporter assay confirmed that miR-486 both targeted the 3'-UTR of SNHG15 and CDK14 and was negatively correlated with their expression levels. In summary, our study conclude that the ectopic overexpression of SNHG15 contribute to the NSCLC tumorigenesis by regulating CDK14 protein via sponging miR-486, providing a novel insight for NSCLC pathogenesis and potential therapeutic strategy for NSCLC patients.

INTRODUCTION

Non‐small cell lung cancer (NSCLC) is one of the leading causes of cancer‐related mortality worldwide, accounting for nearly 80% of all lung cancer cases (Guo, Liu et al., 2017; Zhang, Gao, & He, 2017). In spite of the therapeutic advances in diagnosis and clinical treatment for NSCLC in decades, the 5‐year overall survival rate of NSCLC patients is still approximately 15% and the recurrence rate remains high (Fassina, Cappellesso, & Fassan, 2011; Ma, Yang, Tu, & Hu, 2017; Yang et al., 2017). The pathogenesis and mechanism of NSCLC are complicated, which cause the difficulties for clinical treatment strategies and prognosis. Therefore, it is urgently necessary to identify specific prognostic factors and investigate the underlying molecular mechanisms involved in the metastasis and progression of NSCLC.

In decades, the role of non‐coding RNAs (ncRNAs), including small ncRNAs and long ncRNAs (lncRNAs, more than 200 nt), has been fully recognized, rather than transcription noise (Murugan, Munirajan, & Alzahrani, 2017; Sherstyuk, Medvedev, & Zakian, 2017). It has been verified that lncRNAs function as important regulators in multiple molecular regulation, involving transcription and post‐transcription regulation, and modification (Alvarez‐Dominguez & Lodish, 2017; Marchese, Raimondi, & Huarte, 2017). For example, lncRNA GAS5 was downregulated in NSCLC tissues and cells and was negatively correlated with miR‐23a expression to alleviate the NSCLC tumorigenesis (Mei et al., 2017). For another example, lncRNA MIAT is upregulated in NSCLC and the overexpression was associated with advanced tumor stage, promoting non‐small cell lung cancer proliferation and metastasis through MMP9 activation (Lai et al., 2017).

LncRNA SNHG15 has been reported as a oncogenic gene in tumors to aggravate carcinogenesis through epigenetic regulation (Ma, Xue et al., 2017; Ma, Huang et al., 2017c). For example, SNHG15 promotes breast cancer proliferation, migration, and invasion by sponging miR‐211‐3p (Kong & Qiu, 2018). In osteosarcoma cells, SNHG15 contributes to proliferation, invasion, and autophagy by sponging miR‐141 (Liu, Hou, Liu, & Zheng, 2017). In the present study, our team investigated the potential roles of lncRNA SNHG15 in the tumorigenesis and progression of NSCLC. Results revealed that SNHG15 was up‐regulated in NSCLC tissue and cells and accelerated the NSCLC proliferation and progression through promoting CDK14 via miR‐486. Our results illuminate a novel molecular mechanism and provide an effective treatment strategy for NSCLC.

MATERIALS AND METHODS

Tissue samples

All the NSCLC tissue samples (35 cases) were obtained from Jinshan Hospital affiliated to Fudan University between Jul 2015 and Nov 2016. The tissue samples were rapidly frozen at −80 °C for using. The study was approved by the Ethics Committee of Jinshan Hospital affiliated to Fudan University. Written informed consents were obtained from each participant.

Cell lines and culture

The NSCLC cell lines (A549, H460, SK‐MES‐1, and Calu‐3) were purchased from the American Type Culture Collection (ATCC, Manassas, VA) and cultured in Dulbecco's Modified Eagle's medium (DMEM, Invitrogen, Carlsbad, CA). The normal human bronchial epithelial cells (NHBE) and HEK‐293T cells were purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China) and cultured in RPMI‐1640 medium (Gibco, Waltham, MA) supplemented with 10% FBS (Gibco), 100 U/ml penicillin and 100 µg/ml streptomycin. Cells were cultured in DMEM medium (Invitrogen). All cells were cultured in a 5% CO2 humidified atmosphere at 37 °C.

Transfection

All these interfering RNAs and sequences were chemically synthesized by GenePharma (Shanghai, China), including siRNA against SNHG15 (si‐SNHG15), short‐hairpin RNA plasmid directly targeting SNHG15 (shSNHG15), miR‐486 inhibitor, control miRNA (miR‐NC). Oligonucleotide transfection into NSCLC cells were performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. After transfection of 48 hr, the cells were collected for further experiment.

RNA extraction and quantitative real‐time polymerase chain reaction (qRT‐PCR)

Total RNA was extracted from NSCLC tissues and cells using TRIzol reagent (TaKaRa, Dalian, China). The concentration and purity of RNA were measured at 260/280 nm using NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA). The reverse transcription (RT) reactions were performed using a Prime Script™ RT Reagent Kit (Takara). The specific primers used for RT‐PCR were as following: SNHG15, 5′‐TTGTGAAGCCCAGTGAAAGTACTGC‐3′ (forward), 5′‐TTCACTGTGGAGACTGTCGTTGGT‐3′ (reverse); miR‐486, 5′‐GTGAGTGGAACAATGTGTGG‐3′ (forward), 5′‐AAAGCAAACGATGCCAAGAC‐3′ (reverse); GAPDH, 5′‐AGCCACATCGCTCAGACA‐3′ (forward), 5′‐TGGACTCCACGACGTACT‐3′ (reverse). The results were normalized to GAPDH to obtain the comparative cycle threshold (Ct) (2−ΔΔCt) values, and then converted to fold changes.

Proliferation ability assay

Colony formation assay and CCK‐8 assay were performed to test the proliferation ability of NSCLC cells. For colony formation assay, the transfected cells (A549 and H460) were seeded into a six‐well plates (1000 cells/well) and incubated at 37 °C in 5% CO2 for 10 days. Last, cells were fixed with 4% paraformaldehyde and stained with crystal violet for number counting. Cell Counting Kit‐8 (CCK‐8, Dojindo, Tokyo, Japan) assay was performed according to the manufacturer's illustration. The absorbance of CRC cells was measured at 450 nm at pointed time. CCK‐8 experiments were performed in triplicate.

Flow cytometry apoptosis and cycle analysis

Apoptosis and cell cycle were analyzed using flow cytometry. Briefly, A549 and H460 cells were treated with 400 µl binding buffer, 5 µl Annexin V‐FITC, and 5 µl propidium iodide (PI) using Annexin V‐FITC Apoptosis Detection Kit (Invitrogen). And then cells were seeded into six‐well plates and the apoptosis was measured using BioVision Annexin V‐FITC reagent kit (Sigma–Aldrich, Louis, MO). Finally, the apoptotic rate was measured using a Beckman–Coulter CyAN ADP Analyzer (Beckman Coulter, Inc. Kraemer Boulevard Brea, CA). For cycle analysis, A549 and H460 cells were resuspended using PBS containing 70% ethanol, Rnase A (0.5 mg/ml) and propidium iodide (0.1 mg/ml) and measured using flow cytometry.

Western blot analysis

Total protein was lysed using Radio Immunoprecipitation Assay (RIPA) buffer (Thermo Scientific, Rockford, IL) added with protease inhibitor cocktail (Sigma–Aldrich). Then, protein was then separated on SDS–PAGE gel (8%) and transferred to PVDF membrane (Millipore, Billerica, MA). PVDF membrane was blocked with 5% non‐fat milk for 2 hr at room temperature and incubated with primary antibodies (anti‐CDK14, 1:1000, Abcam, Cambridge, MA) at 4 °C overnight. Next, PVDF membrane was incubated with horseradish peroxidase (HRP) secondary antibody for 1 hr at room temperature. Finally, the protein expression was detected using the enhanced chemiluminescence (ECL) substrate kit (Amersham Biosciences, Inc.) using chemiluminescence detection system.

Luciferase reporter assay

The SNHG15 fragments, containing the wildtype (SNHG15‐WT) or mutant type (SNHG15‐Mut) binding sites with miR‐486 were synthesized by PCR and cloned into the pmirGLO‐basic luciferase reporter vector (Promega, Madison, WI). Briefly, NSCLC cells were co‐transfected with combined luciferase reporter vectors (100 ng), Renilla luciferase vector (20 ng) (Promega), and miR‐486 mimics (100 nM) or miR‐NC using Lipofectamine 2000 (Invitrogen). At 48 hr of transfection, the luciferase activity was tested using the Dual‐Luciferase Reporter assay system (Promega) normalized to Renilla luciferase activity.

Tumor xenograft mice assay

BALB/c nude mice (10 mice, 8–10 g, 4 week) were maintained under pathogen free condition. 1.0 × 107 A549 cells stably transfected with lentivirus‐mediated sh‐SNHG15 or sh‐NC were respectively injected subcutaneously into the right side of the posterior flank of nude mice according to the previous description (Li et al., 2018; Liu, Li et al., 2017). After the injection, tumor volume was measured by a single individual with a digital caliper every 3 days were calculated using the formula 0.5× length × width × width. After the mice were sacrificed, the tumors were extracted and weighted. All the animal experimental procedures were approved by the Animal Experimental Ethics Committee of Jinshan Hospital affiliated to Fudan University.

Statistical analysis

Data were presented as mean ± standard deviation (SD). All results were analyzed with GraphPad Prism 6 using student's t‐tests or one‐way ANOVA. p < 0.05 was considered to indicate statistical significance.

RESULTS

LncRNA SNHG15 was up‐regulated in NSCLC tissue and cells and indicated the poor prognosis

To investigate the expression levels of lncRNA SNHG15, RT‐PCR was performed in the collected NSCLC tissue samples. The clinicopathological data was shown in Table 1. Results showed that lncRNA SNHG15 expression was markedly up‐regulated in 35 cases of NSCLC tissue compared with adjacent non‐tumor tissue samples (Figure 1a). According to TNM pathological grades, SNHG15 expression was significantly high‐expressed in III stage compared with I–II stage (Figure 1b). Meanwhile, SNHG15 expression was significantly over‐expressed in NSCLC cell lines (A549, H460, SK‐MES‐1, and Calu‐3) compared with normal human bronchial epithelial cells (NHBE) (Figure 1c). According to the median value of SNHG15 expression, NSCLC patients were divided into high expression level group and low expression level group (Figure 1d). Kaplan–Meier survival curves and log‐rank tests were performed to evaluate survival rate and results showed that the NSCLC patients with high SNHG15 expression had poor prognosis compared with that of low expression group (Figure 1e). Overall, the data concluded that lncRNA SNHG15 was up‐regulated in NSCLC tissue and cells, and indicated the poor prognosis of NSCLC patients.

Table 1

Relationship between SNHG15 expression and clinicopathological characteristics of 35 cases of NSCLC patients

		SNHG15 expression
Variable	N	Low (15)	High (20)	p
Gender
Male	25	9	16	0.5974
Female	10	6	4
Age (years)
≥60	22	10	12	0.3983
<60	13	5	8
Smoking status
Smoking	25	9	16	0.0129 *
No‐smoking	10	6	4
Differentiation
Well	17	7	10	0.4409
Moderate/poor	18	8	10
TNM
I‐II	21	9	12	0.0228 *
III	14	6	8
Lymphonode metastasis
No	13	7	6	0.0059 *
Yes	22	8	14

p < 0.05 represents statistical difference.

Figure 1

LncRNA SNHG15 was up‐regulated in NSCLC tissue and cells and indicated the poor prognosis. (a) RT‐PCR showed the SNHG15 expression in 35 cases of NSCLC tissue and adjacent noncancerous tissue. (b) SNHG15 expression in I–II stage and III stage of NSCLC tissue according to TNM pathological grades. (c) SNHG15 expression in NSCLC cell lines (A549, H460, SK‐MES‐1, and Calu‐3) and normal human bronchial epithelial cells (NHBE). (d) The distinction of NSCLC patients samples according to the SNHG15 expression (high/low). (e) Kaplan–Meier survival curves and log‐rank tests evaluated the survival rate of NSCLC patients with high or low SNHG15 expression. *p < 0.05, **p < 0.01 presented significant difference

LncRNA SNHG15 silencing suppressed the proliferation of NSCLC cells in vitro

Prior experiments showed that lncRNA SNHG15 was significantly up‐regulated in NSCLC tissue and cells. In further experiments, loss‐of‐function assays were performed to test the role of SNHG15 on the proliferation of NSCLC cells in vitro. SNHG15 expression was significantly down‐regulated in A549 and H460 cell lines transfected with si‐SNHG15‐1 (Figure 2a). CCK‐8 assay showed that SNHG15 silencing suppressed the proliferative ability of NSCLC cells in vitro compared with empty control cells (Figures 2b and 2c). Colony formation assay showed that SNHG15 silencing decreased the clone number in A549 and H460 cells compared with empty control cells (Figures 2d and 2e). In summary, our results showed that lncRNA SNHG15 silencing suppressed the proliferation of NSCLC cells in vitro.

Figure 2

LncRNA SNHG15 silencing suppressed the proliferation of NSCLC cells in vitro. (a) RT‐PCR showed the SNHG15 expression in A549 and H460 cells transfected with three siRNA targeting SNHG15. (b,c) CCK‐8 assay showed the absorbance at 450 nm of A549 and H460 cells. (d,e) Colony formation assay showed the clone number in A549 and H460 cells respectively transfected with si‐SNHG15 and si‐NC. *p < 0.05, **p < 0.01 presented significant difference

LncRNA SNHG15 silencing induced cycle arrest at G0/G1 phase and suppressed the apoptosis of NSCLC cells

To further explore the role of SNHG15 silencing on NSCLC cells phenotype, flow cytometry, and Western blot analysis were performed. Flow cytometry showed that SNHG15 silencing accelerated the apoptosis of A549 and H460 cells compared with empty control transfected cells (Figures 3a and 3b). Besides, flow cytometry cycle analysis revealed that SNHG15 silencing induced the cycle arrest at G0/G1 phase compared with empty control transfected cells (Figures 3c and 3d). Western blot showed that SNHG15 silencing decreased the CDK14 protein expression in A549 and H460 cells (Figures 3e and 3f). In summary, our results showed that SNHG15 silencing induced the cycle arrest at G0/G1 phase and suppressed the apoptosis of NSCLC cells, besides, SNHG15 silencing also decreased the CDK14 protein.

Figure 3

LncRNA SNHG15 silencing induced cycle arrest at G0/G1 phase and suppressed the apoptosis of NSCLC cells. (a,b) Flow cytometry revealed the apoptosis of A549 and H460 cells transfected with si‐NC or si‐SNHG15. (c,d) Flow cytometry cycle analysis revealed the cells distribution of A549 and H460 cells transfected with si‐NC or si‐SNHG15. (e,f) Western blot showed the CDK14 protein expression in A549 and H460 cells. *p < 0.05, **p < 0.01 presented significant difference

SNHG15 silencing inhibited the tumogrowth of NSCLC cells in vivo

To validate the role of SNHG15 on NSCLC tumor growth, xenograft mice assay was performed using A549 cells in vivo (Figure 4a). Results showed that SNHG15 silencing in A549 cells transfected with sh‐SNHG15 decreased the tumor volume of NSCLC cells compared with empty vector transfection (sh‐NC) (Figure 4b). Moreover, SNHG15 silencing down‐regulated the tumor weight of NSCLC cells compared with empty vector transfection (Figure 4c). Western blots analysis showed that CDK14 protein expression was decreased in SNHG15 silencing group compared with control group (Figures 4d and 4e). Overall, xenograft assay in vivo showed that SNHG15 silencing inhibited the tumor growth of NSCLC cells, and down‐regulated the CDK14 protein expression.

Figure 4

SNHG15 silencing inhibited the tumor growth of NSCLC cells in vivo. (a) The images of xenograft assay using A549 cells. (b) Tumor volume was measured after subcutaneous injection every 3 days. (c) Tumor weight was measured after mice sacrifice. (d) Blot images of CDK14 protein. (e) Quantitative data of CDK14 protein expression. Data were expressed as mean ± SD. *p < 0.05, **p < 0.01 represents statistically difference

SNHG15 positively regulated CDK14 expression via sponging miR‐486

Previous study had revealed that SNHG15 positively regulated CDK14 protein expression. In follow‐up assay, we investigate the underlying regulatory mechanism within SNHG15 and CDK14. Bioinformatics prediction tools demonstrated that miR‐486 targeted 3′ untranslated regions (3′‐UTR) of SNHG15 with eight complementary binding sites (Figure 5a). Luciferase reporter assay revealed that miR‐486 targeted SNHG15 3′‐UTR with molecular binding (Figure 5b). In NSCLC tissue samples, miR‐486 expression was decreased compared with adjacent non‐tumor tissue (Figure 5c). Then, bioinformatics prediction tools illustrated that miR‐486 targeted 3′‐UTR of CDK14 with complementary binding sites (Figure 5d). Luciferase reporter assay validated the molecular interaction within miR‐486 and CDK14 (Figure 5e). In A549 cells transfected with si‐SNHG15, miR‐486 expression was significantly increased, while CDK14 mRNA expression was decreased compared with si‐NC transfection (Figure 5f). In A549 cells transfected with miR‐486 inhibitor, CDK14 mRNA, and SNHG15 expression levels were significantly increased compared with control transfection. However, in A549 cells transfected miR‐486 mimics, CDK14 mRNA, and SNHG15 expression levels were significantly decreased (Figure 5g). Western blot showed that miR‐486 inhibitor increased the CDK14 protein expression, and the co‐transfection of miR‐486 inhibitor and si‐SNHG15 abrogated the CDK14 expression (Figures 5h and 5i). Overall, results concluded that SNHG15 positively regulated CDK14 expression via negatively sponging miR‐486, suggesting the SNHG15/miR‐486/CDK14 regulation.

Figure 5

SNHG15 positively regulated CDK14 expression via sponging miR‐486. (a) Schematic diagram showed the 3′‐UTR of SNHG15 wild‐type and mutant and miR‐486. (b) Luciferase reporter assay showed the luciferase activity of the combination of miR‐486 and SNHG15 wild‐type/mutant. (c) miR‐486 expression in NSCLC tissue samples compared with adjacent non‐tumor tissue. (d) Schematic diagram showed the binding within 3′‐UTR of CDK14 wild‐type and mutant and miR‐486. (e) Luciferase reporter assay showed the luciferase activity of the combination of miR‐486 and CDK14 wild‐type/mutant. (f) miR‐486 expression and CDK14 mRNA expression Iin A549 cells transfected with si‐SNHG15. (g) RT‐PCR showed CDK14 mRNA and SNHG15 expression levels in A549 cells transfected with miR‐486 inhibitor or miR‐486 mimics. (h,i) Western blot showed the CDK14 protein expression in A549 cells co‐transfected with miR‐486 inhibitor and si‐SNHG15. Data were expressed as mean ± SD. *p < 0.05, **p < 0.01 represents statistically difference

DISCUSSION

Emerging evidence have shown that lncRNAs function as oncogenic or suppressor genes in multiple cancers, involving in transcription and post‐transcription regulation (Guo, Ma et al., 2017; He et al., 2017). In present study, our team performed series of experiments to investigate the biological role of lncRNA SNHG15 on non‐small‐cell lung cancer (NSCLC) tumorigenesis.

The tumorigenesis of NSCLC is a complex pathophysiological processes and numerous factors participate in the pathology (Yang et al., 2017; Zhang, Wu et al., 2017b). Long non‐coding RNAs (lncRNAs) are a type of non‐coding RNA transcriptome without protein coding ability. The vital role of lncRNAs has been adequately verified in NSCLC tumorigenesis correlated with multiple molecular levels, including differentiation, proliferation, apoptosis, and metastasis. In this study, our results found that lncRNA SNHG15 expression was markedly up‐regulated in NSCLC tissue samples compared with adjacent non‐tumor tissue samples, besides, the high level of SNHG15 indicated poor prognosis of NSCLC patients. Cellular experiments in vitro concluded that SNHG15 silencing suppressed the proliferation of NSCLC cells. In general, the ectopic overexpression of lncRNAs indicates the negative or oncogenic role of lncRNAs in tumorigenesis. For example, lncRNA BLACAT1 (bladder cancer associated transcript 1) is significantly upregulated in NSCLC tissues compared with adjacent normal tissues and BLACAT1 suppresses proliferation, migration, invasion, and induce G0/G1 phase arrest via sponging miR‐144 (Ye, Liu, & Zheng, 2017).

On the other hand, we found that SNHG15 silencing induced the cycle arrest at G0/G1 phase and suppressed the apoptosis of NSCLC cells, besides, decreased the CDK14 protein. CDK14 acts as one of the important cell cycle‐regulation related genes (Li et al., 2014; Pollack et al., 2015). Meanwhile, SNHG15 regulated the apoptosis and cell‐cycle of NSCLC cells, suggesting the potential interaction within SNHG15 and cycle modulation. CDK14 protein expression is positively correlated with that of SNHG15. The cycle arrest at G0/G1 phase might induce the cellular process of NSCLC cells and accelerate the apoptosis, casing the blocking of NSCLC progression.

LncRNAs is a type of non‐coding RNA without protein coding ability (Song, Sakurai, Shiromoto, & Nishikura, 2016; Wei & Wang, 2015). Up to now, the main regulation mode of lncRNAs is the miRNA “sponge,” indicating the regulation of lncRNAs on functional proteins via sponging miRNAs (Lorenzen & Thum, 2016; Thomson & Dinger, 2016). In our study, bioinformatics online tools and luciferase assay found that miR‐486 targeted the 3′‐UTR of lncRNA SNHG15 and SNHG15 function as molecular “sponge” for miR‐486. Meanwhile, our results also illustrated that miR‐486 targeted CDK14 3′‐UTR. Therefore, we conclude the regulation of SNHG15/miR‐486/CDK14 in the NSCLC tumorigenesis. The lncRNA/miRNA/mRNA regulatory pathway has been verified to be an important modulation in NSCLC (Guo, Wang, Ren, & Han, 2018; Jiang et al., 2017; Wang, Chen, Ma, & Li, 2017). For example, lncRNA MALAT1 is significantly upregulated in five human NSCLC cells and bioinformatic analysis predicts the correlation between miR‐124 and MALAT1, and STAT3 is found to be a novel mRNA target of miR‐124 (Li, Mei, & Hu, 2017).

In summary, our results reveal that lncRNA SNHG15 is over‐expressed in NSCLC tissue and cell lines, and indicates the poor prognosis. Furthermore, we find that SNHG15 promotes CDK14 protein expression through sponging miR‐486 to modulate the NSCLC tumorigenesis, providing a novel insight for NSCLC pathogenesis and potential therapeutic strategy for NSCLC patients.

CONFLICTS OF INTEREST

All authors declare no conflicts of interest.