key: cord-0028777-cmymguxf
authors: Zhong, Qingmu; Chen, Yifeng; Chen, Zilong
title: LncRNA MINCR regulates irradiation resistance in nasopharyngeal carcinoma cells via the microRNA-223/ZEB1 axis
date: 2019-11-24
journal: Cell Cycle
DOI: 10.1080/15384101.2019.1692176
sha: c7db932a129b3943a08d386066f964664e7341a9
doc_id: 28777
cord_uid: cmymguxf

Emerging evidence suggests long non-coding RNA (lncRNA) could sponge microRNAs (miRs) and monitor gene expression. In this study, we intended to search the network involving lncRNA MINCR/miR-223/ZEB1 in nasopharyngeal carcinoma (NPC) cell radiosensitivity. MINCR expression in NPC tissues, precancerous lesions and chronic nasopharyngeal mucosal inflammation tissues, and in NP460, CNE2 and CNE2R cells was detected. The associations between MINCR expression and prognosis and radiotherapy efficacy of NPC patients were evaluated. The interactions among MINCR, miR-223 and ZEB1 were verified via dual luciferase reporter gene assay, RNA pull-down and FISH assays. The gain- and loss-of-functions were performed to explore their effects on NPC cell viability, apoptosis and radiosensitivity. Levels of MINCR, miR-223, ZEB1, and AKT/PI3K-related proteins were detected after different treatments. An in vivo analysis was carried out in nude mice. Consequently, MINCR was upregulated in NPC, and linked with worse prognosis and radiotherapy efficacy. MINCR intervention weakened NPC cell radioresistance. MINCR sponged miR-223 to regulate ZEB1. Inactivating AKT eliminated the increased radioresistance of CNE2 cells induced by overexpressing MINCR. Briefly, MINCR diminished NPC cell radiosensitivity by sponging miR-223, increasing ZEB1 and activating the AKT/PI3K axis. This study may offer novel insight for NPC treatment.

Nasopharyngeal carcinoma (NPC) is a malignant neoplasm occurring in the top and lateral walls of the nasopharyngeal cavity [1] . As a kind of malignant epithelial carcinoma in the head and neck, NPC has strong ability of local invasion and early distant metastasis, which is the major cause for poor prognosis in NPC patients at advanced period [2] . NPC patients usually manifest nasal and aural symptoms, including nasopharyngeal mass, dysfunction of Eustachian tube, cranial nerve palsy and cervical masses, sometimes complete nasal obstruction [3] Epstein-Barr virus infection is a well-recognized risk factor for NPC development, and other co-factors including dietary nitrosamine consumption, long-term exposure to wood dusts or chemical carcinogens, cigarette smoking and genetic susceptibility are also of great importance [4] . Although the management and treatment of NPC has been improved by imaging and advanced radiotherapy techniques, the prognosis of NPC patients is still unsatisfactory in the past decades mainly due to radioresistance [5] . Therefore, the future research for NPC therapeutic methods should focus on cell radioresistance.

Long non-coding RNAs (lncRNAs) are regulatory RNAs that exert biochemical functions, including in the pathological mechanism of diseases, cell differentiation and proliferation, cancer metastasis and development, and especially NPC cell radioresistance [6] . MINCR is a Myc-induced lncRNA that regulates Myc transcription networks in Burkitt lymphoma cells, Myc-positive lymphoma and pancreatic ductal adenocarcinomas, and MINCR is highly expressed in brain, prostate, and testis [7] . At the post-transcriptional level, lncRNAs compete with microRNAs (miRs) to act as a "sponge" and that lncRNA functioning as a competing endogenous RNA (ceRNA) is a newly proposed hypothesis that has drawn increased attention [8] . In this study, lncRNA MINCR was found to bind to miR-223. Interestingly, overexpressed miR-223 suppresses FBXW7 expression in esophageal squamous cell carcinoma cells, then leading to abundant production of c-Myc and c-Jun proteins [9] . Recent evidence demonstrates that miR-223 is effective in regulating chemotherapeutic drug sensitivity in non-small cell lung cancer [10] . In NPC, miRs play pivotal regulatory roles in cell growth, proliferation and radiosensitivity [11, 12] . Importantly, differential expression of miR-223 is reported in serum of NPC patients [13] . Such statements encourage us to suppose the ceRNA network between MINCR and miR-223 in NPC cell radioresistance with potential signaling pathway.

The study was performed with the approval of the Ethics Committee of First Hospital of Quanzhou Affiliated to Fujian Medical University. All participating patients subscribed the informed consent. Significant efforts were made to minimize the number of animals and their pains.

Tissue biopsy was obtained from 49 NPC patients, 16 patients with precarcinoma lesion (PL) and 34 patients with nasopharyngeal mucosa chronic inflammation (NMCI). They were admitted into the Department of Pathology in First Hospital of Quanzhou Affiliated to Fujian Medical University between January 2012 to January 2014. No patients had a medical history of other malignant tumors, radiotherapy or chemotherapy. Each NPC patient was given 10 Gy irradiation every week. Based on the Response Evaluation Criteria in Solid Tumors, NPC patients were assigned into responders (n = 21) and non-responders (n = 28) to irradiation. Their clinical stages were defined according to the 2002 AJCC/UICC staging classifications [14] . Of the 49 cases, three were at stage I, 13 at stage II, 16 at stage III, and 17 at stage IV. The overall survival (OS) was the duration from the initial date of treatment to date of death. The follow-up lasted for 60 months and was ended in case of tumor reoccurrence or death; otherwise, OS time was recorded to the last follow-up.

NPC cell lines, CNE2 and CNE2R, provided by American Type Culture Collection (ATCC, Manassas, VA, USA) were maintained in Roswell Park Memorial Institute (RPMI)-1640 medium containing 10% newborn calf serum at 37°C in a humidified condition with 5% CO 2 and 95% air. The protein kinase B (AKT) inhibitor, GSK690693, was obtained from Selleck Chemicals (Houston, TX, USA).

Short interfering RNAs (siRNAs) (100 nM) specifically targeting MINCR were designed to construct MINCR-depleted CNE2R cells. The corresponding scramble siRNAs (100 nM) were delivered to NPC cells to serve as negative control (NC). The vectors containing MINCR (50 nM) were introduced for construction of MINCR-overexpressed NPC cells. All plasmids were provided by Shanghai GenePharma Co, Ltd (Shanghai, China). Subsequently, miR-223 mimic (50 nM) and ZEB1 vectors (50 nM) were transfected into MINCR-overexpressed CNE2 cells and MINCR-depleted CNE2R cells, respectively using Lipofectamine TM 2000 (Invitrogen Inc., Carlsbad, CA, USA) as per the instructions, lasting for 48 hours.

Total RNA was extracted from tissues and cells by the method of Trizol (Invitrogen, Carlsbad, CA, USA) and reversely transcribed into cDNA as per the instructions of PrimeScript TM II 1st Strand cDNA Synthesis Kits (Takara, Dalian, China) [15] . RT-qPCR was conducted with the SYBR green Premix Ex Taq II (RR420A, Takara Holdings Inc., Kyoto, Japan) [16] on Applied Biosystems Step One Plus Real-Time PCR System (Applied Biosystems, Carlsbad, CA, USA). Each reaction was run for three times. Data were normalized to the fold change of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and relative expression of MINCR and ZEB1 was determined using the ΔΔCt method. Primer sequences are exhibited in Table 1 .

Cells were resuspended with RPMI-1640 medium (Gibco, Gaithersburg, MD, USA) in Luria-Bertani culture plate (D0110, Beijing Nobleryder Science and Technology Co., Ltd., Beijing, China). A total of 500 cells were seeded in the medium (10 cm) and cultured at 37°C with 5% CO 2 for two weeks. Then cells were exposed to 0 Gy, 4 Gy and 8 Gy irradiation for 2 hours. After that, cells were fixed with 4% paraformaldehyde for 20 minutes, and then stained with crystal violet for 60 minutes. The plates were air dried to count clones with more than 50 cells under a microscope.

Cell proliferation after transfection was examined by a CCK-8 assay. In detail, cells were seeded into 96-well plates with 3 × 10 3 cells per well for 3 days, followed by being exposed to 8 Gy irradiation. Then, cells were incubated with 10 μL CCK-8 solution (5 mg/mL, Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) for 4 hours. Two hours later, the medium in each well was added with 150 μL dimethyl sulfoxide (Sigma-Aldrich) to dissolve the formazan crystals. The optical density value was read at 490 nm using a microplate reader (Bio-Rad, Hercules, CA, USA) and the growth curves were drawn.

An apoptosis assay was performed using the fluorescein isothiocyanate Annexin V Apoptosis Detection Kit (KeyGen, Nanjing, Jiangsu, China) [17] . Propidium iodide was used with Annexin V to determine if cells were viable, apoptotic, or necrotic, and analyzed by flow cytometry (FACScan®) and CellQuest software (both from BD Biosciences, San Jose, CA, USA).

The subcellular localization of MINCR was predicted using lncRNA subcellular localization data (lncatlas.crg.eu.) and verified with fluorescence in situ hybridization (FISH) using Ribo TM lncRNA FISH Probe Mix (Green) (Ribo Biotech, Guangzhou, Guangdong, China). NPC cells (6 × 10 4 /well) were mounted onto slides and fixed in 4% formaldehyde. Slides were pretreated with protease K (2 μg/mL), glycine and acetic anhydride, followed by pre-hybridization for 1 hour at 42°C and hybridization at 42°C using probes (250 μL, 300 ng/mL) against MICNR. Finally, slides were stained with phosphatebuffered saline with Tween-diluted 4ʹ,6-diamidino-2-phenylindole (DAPI). Finally, five random fields acquired from each slide were observed and photographed under a fluorescence microscope (Eclipse Ti microscope, Nikon Instruments, Chiyoda-ku, Tokyo, Japan).

The nuclear and cytoplasmic RNA fractions were isolated according to PARIS TM Kit (Life Technologies, Inc., Gaithersburg, MD, USA). CNE2 cells were resuspended for 5-10 minutes in 500 µL cell fractionation buffer. The cytoplasmic nuclear fractions were separated by centrifugation at 500 g and 4°C for 5 minutes. The supernatant (cytoplasmic fractions) was put into a 2 mL sterile enzyme-free tube and then centrifuged. The pellet (nuclear fraction) was resuspended in 500 µL cell disruption buffer and centrifuged. The cytoplasmic and nuclear fractions were respectively immersed in 500 µL 2 × lysis/ binding solution and centrifuged. Next, the factions were mixed with 500 µL absolute ethanol and then transferred into a filter cartridge, which was subsequently washed with Wash solution 1 and 2/3. Nuclear and cytoplasmic RNA were obtained after elution. MINCR expression was determined by RT-qPCR, with U6 used as the 

Cell lysates were treated with RNase-free DNase I (Sigma-Aldrich) and cultured with a mixture of biotinylated RNA fragments of miR-223 (1 µg) and streptavidin-coated magnetic beads (Sigma-Aldrich) at 4°C for 3 hours. After that, the RNA was extracted from the harvested RNA-RNA complexes for western blot analysis.

Cells were lysed in ice-cold radioimmunoprecipitation buffer containing protease inhibitor phenylmethylsulfonyl fluoride (1 mM). Equal protein samples (50 μg) were run on 10% sodium dodecyl sulfate -polyacrylamide gel electrophoresis (Bio-Rad, Hercules, CA, USA) and then transferred to polyvinylidene fluoride membranes (Amersham Pharmacia, Piscataway, NJ, USA). Next, western blots were probed with antibodies against ZEB1 (1:1,000; ab203829), AKT (1:1000; ab8805), p-AKT (1:10,000, ab81283), phosphoinositide-3-kinase (PI3K, 1:1000, ab191606), p-PI3K (1:1000, ab182651) and β-actin (1:5000; ab227387) (all purchased from Abcam, Cambridge, MA, USA) at 4°C overnight, and then probed with secondary antibody, horseradish peroxidase-labeled goat antirabbit immunoglobulin G (IgG) (1:2,000; A0208; Beyotime Institute of Biotechnology, Shanghai, China). Each sample was repeated for three times. GAPDH served as a reference for normalization. At last, immunoblots were visualized with enhanced chemiluminescence (Amersham Pharmacia) and analyzed with ImageJ V1.48 software (National Institutes of Health, Bethesda, Maryland, USA).

Nude mice (BALB/C nu/nu) (Chinese Academy of Medical Sciences, Beijing, China) were fed with autoclaved water and laboratory rodent chow. The culture medium (100 μL) was mixed with Matrigel (BD Biosciences) containing 3 × 10 6 CNE2R cells which were stably transfected si-MINCR or scramble siRNA, and then transplanted into the flanks of nude mice by subcutaneous injection. By the way, the tumor volume was measured every 5 days for totally 24 days and calculated as follows: Volume (mm 3 ) = (a × b 2 )/2, where "a" indicated the largest diameter and "b" represented the perpendicular diameter. Once tumors reached about 70 mm 3 , the mice were randomly allocated into 3 groups (3 mice/group) and given 2 × 5 Gy irradiation.

Tumor sections at 5 μm from NPC xenografts were stained with anti-p-AKT antibody (1:10,000, ab81283) and anti-Ki67 antibody (1:50, MIB-1, Immunotech, Marseille, France) at 4°C overnight and then reacted with anti-IgG secondary antibody (1:1,000; ab6721, Abcam) for 30 minutes. After that, visualization was performed using 3,3-diaminobezidine (DAB, DA1010, Solarbio, Beijing, China). Finally, 5 fields of view (200 ×) were randomly captured for each replicate under an inverted microscope (Nikon).

SPSS 21.0 (IBM Corp., Armonk, NY, USA) was applied for data analysis. Kolmogorov-Smirnov test showed whether the data were in normal distribution. The results were manifested as mean ± standard deviation. Comparison between the two groups was analyzed by unpaired t test, comparison among multiple groups was analyzed by oneway or two-way analysis of variance (ANOVA), and pairwise comparison after ANOVA was conducted by Sidak's multiple comparisons test or Tukey's multiple comparisons test. The log rank test was applied for post statistical analysis. The p value was obtained by two-tailed tests and p < 0.05 indicated significant difference.

MINCR is reported to contribute to several types of cancers [18] . However, the role of MINCR in NPC tumorigenesis remains unknown. Thus, we analyzed nasopharyngeal mucosal tissues from 34 NMCI patients, nasopharyngeal PL tissues from 16 PL patients, and NPC biopsy tissues from 49 NPC patients. The results displayed highly expressed MINCR in NPC biopsy tissues (Figure 1(a) ). Besides, NPC patients with high MINCR expression exhibited worse prognosis in the follow-up records (Figure 1(b) ). Subsequently, 49 NPC patients, owing to irradiation effectiveness, were split into 2 groups, the resistant and sensitive groups. We analyzed the relationship between radiotherapy efficacy and MINCR level in NPC patients, and found that the higher MINCR level led to worse radiotherapy efficacy in NPC patients (Figure 1(c) ). In order to determine the expression of MINCR and its clinical value in predicting radiotherapy efficacy on NPC patients, we evaluated the radiotherapy efficacy of MINCR on NPC by using receiver operator characteristic (ROC) curve and area under curve (AUC). The results showed that MINCR expression was of better prediction on radiotherapy efficacy in NPC patients (Figure 1(d) ).

CNE2R was confirmed to be radioresistant cells at different dose of irradiation by colony formation assay (all p < 0.05) (Figure 1(e) ). Subsequently, MINCR expression in normal nasopharyngeal epithelial cell line NP460, human NPC cell line CNE2 and radioresistant cell line CNE2R was determined by RT-qPCR. The result pointed out MINCR was notably overexpressed in radioresistant CNE2R cells (Figure 1(f) ).

To further verify the radio-sensitivity of MINCR to NPC cells, CNE2R cells interfering with MINCR expression and CNE2 cells overexpressing MINCR were constructed, and RT-qPCR verified the transfection was successful (all p < 0.05) (Figure 2(b) ). Twelve days after CNE2 and CNE2R cells received irradiation at doses of 0, 4 and 8 Gy, colony formation assay showed that MINCR interference resulted in a decrease in radioresistance of NPC cells, and NPC cell proliferation was decreased further with the increase of irradiation dose (p < 0.05) (Figure 2(b) ). Subsequently, we detected the proliferation of NPC cells after different treatments. The results inferred that interfering MINCR expression promoted NPC cell radiosensitivity (all p < 0.05) (Figure 2(c) ). In addition, flow cytometry showed that silencing MINCR significantly increased apoptotic cells after irradiation (all p < 0.05) (Figure 2(d) ). Briefly, NPC cells with interfering MINCR expression were more sensitive to irradiation.

The protein subcellular localization is of great importance for cell biology, system biology and drug breakthrough [19] . So we first predicted the subcellular localization of lncRNA MINCR through the LncATLAS database, and it turned out that MINCR was mainly in the cytoplasm (Figure 3(a) ). Subsequently, we validated that MINCR was located in the cytoplasm of CNE2 cells by FISH experiments.

The probes targeting MINCR in CNE2 cells were stained in green and the nucleus were stained in blue after DAPI staining (Figure 3(b) ). Additionally, we extracted total RNA from CNE2 cells by fractionation of nuclear and cytoplasmic RNA, and detected the expression of MINCR in cytoplasm and nucleus respectively. The results showed that MINCR mainly existed in cytoplasm (p < 0.05) (Figure 3(c) ).

How MINCR participates in radiotherapy resistance of NPC cells has not been reported. To study this, we first predicted a large number of miRs that bound to MINCR through StarBase [20] . As reported, overexpression of miR-223 can Figure 1 . MINCR is at a high level in NPC and predicts poor prognosis. a. MINCR expression in NMCI patients, PL patients and NPC patients detected by RT-qPCR; b. Relationship between MINCR expression and prognosis in NPC patients analyzed by Kalpan-Meier analysis (patients with high MINCR expression had 40.3 months median survival periods, while patients with low MINCR expression had 54.2 months median survival periods); c. The resistant group patients bore higher MINCR level determined by RT-qPCR; d. ROC analysis for prediction of MINCR level on radiotherapy efficacy (areas under the ROC curve: 0.719; sensitivity: 66.7% and specificity: 71.0%); e. Colony formation assay was performed to affirm CNE2R cells possessed with radioresistance compared to its parental cell CNE2; F. Relative MINCR level in CNE2R cells, CNE2 cells and nasopharyngeal epithial cell NP460 detected by RT-qPCR. Data were described as mean ± standard deviation. Data in panels A, C and E were analyzed by one-way ANOVA, followed by Sidak's multiple comparisons test, or unpaired t test was used. *p < 0.05, **p < 0.01; In panel A, for NMCI patients, n = 34, for PL patients, n = 16 and for NPC patients, n = 49. In panel C, sensitive group contains 21 candidates and resistant group contains 28. In panel E and F, three independent experiments were performed. NPC, nasopharyngeal carcinoma; NMCI, nasopharyngeal mucosa chronic inflammation; PL, precarcinoma lesion; RT-qPCR, real-time quantitative polymerase chain reaction; ANOVA, analysis of variance.

promote radiosensitivity of U87 cells by downregulating Ataxia-telangiectasia mutated (ATM) axis [21] . We suspected that MINCR may reduce NPC cell radiosensitivity by binding to miR-223. Therefore, according to the binding sequence predicted by Starbase, we designed a luciferase reporter plasmid based on cytomegalovirus (CMV), which contained the binding sites of miR-223 mimic and miR-NC with WT-MINCR or MT-MINCR, respectively. The results showed that miR-223 could specifically bind to MINCR (Figure 4(a,b) ). Subsequently, RNA pull-down assay was performed in order to further verify the binding relationship between miR-223 and MINCR. The experimental results were in line with our expectations (all p < 0.05) (Figure 4(c) ). Besides, RT-qPCR revealed that miR-223 expression was negatively correlated with MINCR in 49 NPC tissues (Figure 4(d) ). We further detected miR-223 expression in NPC cells with overexpressed or silenced MINCR, and the results were identical to the above ones (all p < 0.05) (Figure 4(e) ).

To further identify the downstream gene of miR-223 in radiotherapy resistance, we screened the target genes of miR-223 through Starbase and TargetScan websites, among which we focused on ZEB1. ZEB1 gene, as a transcription factor, can promote the metastasis of cancer cells [22] . Moreover, ZEB1 can enhance the radioresistance of cancer cells by promoting epithelial-mesenchymal transition (EMT) of cancer cells [23, 24] . According to the prediction website Starbase, miR-223 and ZEB1 3ʹ-UTR had specific binding sites (all p < 0.05) (Figure 4(f) ). Based on this sequence, we designed a luciferase reporter plasmid based on CMV, which contained the binding sites of miR-223 mimic and miR-NC with WT or MT ZEB1 3ʹ-UTR, respectively. The results showed that miR-223 could specifically bind to ZEB1 (Figure 4(g) ). Subsequently, RNA pull-down assay was performed Relative survival fraction and cell clones in CNE2 and CNE2R cells with silencing or overexpressing MINCR measured by colony formation assay; c. Optical density value of CNE2 and CNE2R cells with silencing or overexpressing MINCR measured by CCK-8 assay; d. Relative apoptosis of CNE2R and CNE2 cells with silencing or overexpressing MINCR measured by flow cytometry. Data were expressed as mean ± standard deviation. Data in panels A and D were analyzed with one-way ANOVA, followed by Sidak's multiple comparisons test, while data in panels B and C were analyzed with two-way ANOVA and Tukey's multiple comparisons test. *, p < 0.05, **, p < 0.01. Three independent experiments were performed. NPC, nasopharyngeal carcinoma; siRNA, short interfering RNA; RT-qPCR, real-time quantitative polymerase chain reaction; CCK-8, cell counting kit-8; NC, negative control; ANOVA, analysis of variance.

to further verify the binding relationship between miR-223 and ZEB1. The experimental results were in line with our expectations (all p < 0.05) (Figure 4 (h)). Besides, RT-qPCR revealed that ZEB1 expression was positively correlated with MINCR in 49 NPC tissues (Figure 4(i) ). We further detected ZEB1 expression in NPC cells with overexpressed or silenced MINCR, and the results were consistent with the above ones (all p < 0.05) (Figure 4(e) ).

To further confirm the roles of miR-223 and ZEB1 in radioresistance of NPC cells, we constructed the miR-223 mimic, miR-NC and ZEB1 overexpression vectors and empty plasmids, and transfected them into CNE2 cells with overexpressed MINCR and CNE2R cells with silenced MINCR, respectively. The expression of miR-223 and ZEB1 in CNE2 and CNE2R cells was detected by RT-qPCR, and the transfection was confirmed to be successful (all p < 0.05) (Figure 5(a) ).

After 6 Gy doses of radiation, the proliferation and apoptosis of CNE2 and CNE2R were detected. The results suggested overexpression of miR-223 decreased the radioresistance induced by MINCR upregulation, while ZEB1 overexpression in MINCRsilenced CNE2R cells enhanced the radioresistance (all p < 0.05) ( Figure 5(b-d) ). Taken together, miR-223 attenuated NPC cell irradiation resistance.

In 2013, Chen et al reported that ZEB1 was positively correlated with p-AKT expression in NPC patients, and in their in vitro experiments, the sensitivity of NPC cells was significantly improved by adding GSK690693, an AKT inhibitor [25] . In light of this, we detected levels of p-AKT and Luciferase reporter plasmid containing MINCR-WT or MINCR-MT was transfected into CNE2 cells together with miR-223 in parallel with miR-NC plasmid vector; c. The enrichment of miR-223 on MINCR was detected by RNA pull-down assay, relative to antisense-oligos; d. Plot analysis of the relationship between MINCR and miR-223 in the 49 NPC patients; e. Relative expression of miR-223 in NPC cells with overexpressed or silenced MINCR determined by RT-qPCR; f. miR-223 targeting site in ZEB1 3'UTR predicted by TargetScan; g. Luciferase reporter plasmid containing ZEB1-WT or ZEB1-MT was transfected into CNE2 cells together with miR-223 in parallel with miR-NC plasmid vector; h. The enrichment of miR-223 on ZEB1 was detected by RNA pull down assay, relative to antisense-oligos; i. Plot analysis of the relationship between MINCR and ZEB1 in the 49 NPC patients; j. Relative ZEB1 mRNA expression in NPC cells with overexpressed or silenced MINCR determined by RT-qPCR; K. Relative ZEB1 protein level in NPC cells with overexpressed or silenced MINCR determined by western blot analysis. The level of miR-223 was normalized to U6 while the ZEB1 mRNA and MINCR level was normalized to GAPDH. In panels B, D, F and H, all experiments were performed for three times, and one-way ANOVA was used to determine statistical significance. In panels C and G, Pearson's correlation coefficient test was utilized. n = 49. Relative to the empty vector group, *, p < 0.05; compared with the mock group, # , p < 0.05. miR-223, microRNA-223; WT, wild type; MT, mutant type; NPC, nasopharyngeal carcinoma; RT-qPCR, real-time quantitative polymerase chain reaction; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ANOVA, analysis of variance.

p-PI3K in CNE2 and CNE2R cells in each group. The results showed that MINCR enhanced the function of ZEB1 and activated the AKT/PI3K signaling pathway (all p < 0.05) (Figure 6(a) ) through competitively binding to miR-223. Subsequently, we added AKT specific inhibitor GSK690693 to CNE2 cells overexpressed MINCR to perform functional rescue experiments. The results showed that inhibiting AKT activation partially eliminated the increased radioresistance of CNE2 cells induced by overexpressed MINCR (all p < 0.05) (Figure 6(b-d) ). To sum up, MINCR enhanced ZEB1 function and activated the AKT/ PI3K signaling pathway, thus increasing NPC cell radioresistance.

Subsequently, in order to further determine the effect of MINCR on NPC cell growth in vivo, CNE2R cells in nude mice and CNE2R xenograft tumor models stably transfected with MINCR-siRNA were constructed. After radiotherapy, the growth rate and weight of MINCR-siRNA tumors decreased significantly (all p < 0.05) (Figure 7(a) ). Moreover, the immunohistochemical results of p-AKT and Ki67 showed that the contents of p-AKT and Ki67 decreased significantly after interfering MINCR expression (all p < 0.05) (Figure 7(b-c) ). 

Although chemotherapy and radiotherapy are widely used in single interventions or in combination with other anticancer drugs for NPC, drug resistance is still a leading impediment to successful treatment and ultimately causes recurrence and poor prognosis [1, 12] . It has been well established for the importance of the ceRNA network between lncRNA-mRNA interactions in tumorigenesis of cancers [26] . Inspired by that finding, we evaluated the ceRNA regulatory network between MINCR and miR-223 in NPC biological process with the potential signaling pathway. As expected, we found that MINCR sponged miR-223 to increase ZEB1 expression and activate the AKT/PI3K axis, and thus diminishing NPC cell radiosensitivity.

We firstly observed MINCR was highly expressed in NPC tissues and radioresistant CNE2R cells, and NPC patients with high MINCR expression had worse prognosis and worse radiotherapy efficacy. There was no study about the effects of MINCR on NPC, but several researches on the role of MINCR in other cancers could be referred. For instance, upregulation of MINCR in gallbladder cancer was markedly relevant to larger tumor sizes, lymph node metastasis, and shorter overall survival time [18] . Besides, MINCR interference resulted in decreased proliferation and radioresistance of NPC cells, and increased apoptotic cells after irradiation. MINCR knockdown is bound up with impaired cell cycle progression and reduced cellular proliferation [7] . MINCR depletion suppressed cell growth and invasion and stimulated cell apoptosis, by inhibiting the EMT in oral squamous cell carcinoma cells, and interrelated with lower TNM stage and less distant metastasis [27] . Since MINCR is a newly discovered lncRNA, there is little research about its mechanism in cancer cell radioresistance. But the carcinogenic role of other lncRNA in NPC has been widely reported. For example, MALAT1 depletion could reinforce NPC cell radiosensitivity by working as a ceRNA to modulate slug expression through competition for miR-1 [6] .

More importantly, we found the beneficial effects of MINCR interference on reduced NPC cell radioresistance were achieved via miR-223, ZEB1 and the AKT/ PI3K axis. Specifically, miR-223 could bind to MINCR, and it was negatively correlated with MINCR in NPC tissues. The ceRNA network involving MINCR was also discovered in gallbladder cancer cell, in which MINCR could act as a sponge for miR-26a-5p to regulate EZH2 [18] . Additionally, miR-223 could bind to ZEB1, and ZEB1 expression was positively correlated with MINCR in NPC tissues. Figure 6 . MINCR promoted NPC cell autophagy via AKT/PI3K signaling pathway activation. a. Western blot analysis was utilized for determining AKT/PI3K pathway content in CNE2 and CNE2R cell. Then, CNE2/Oe-MINCR was treated with AKT inhibitor GSK690693; b. Relative survival fraction and cell clones in CNE2 cells with overexpressing MINCR and AKT inhibitor detected by colony formation assay; c. Optical density value of CNE2 cells with overexpressing MINCR and AKT inhibitor detected by CCK-8 assay; d. Relative apoptosis in CNE2 cells with overexpressing MINCR and AKT inhibitor detected by flow cytometry. Data were manifested as mean ± standard deviation. Data in panels D were analyzed with one-way ANOVA and Sidak's multiple comparisons test, while data in panel A, B and C were analyzed with two-way ANOVA and Tukey's multiple comparisons test. *, p < 0.05, **, p < 0.01. Three independent experiments were performed. miR-223, microRNA-223; AKT, protein kinase B; PI3K, phosphoinositide-3-kinase; NPC, nasopharyngeal carcinoma; Oe, overexpression; CCK-8, cell counting kit-8; ANOVA, analysis of variance.

Extrogenous miR-223 in CNE2 cells would decrease the ability of colony formation, migration and invasion by reducing its another target gene MAFB [28] . miR-223 overexpression attenuated NPC cell irradiation resistance. Consistently, downregulation of miR-223 was observed in serum of NPC patients in a prior study, serving as a tumor suppressor gene [13] . Similarly, miR-223-3p overexpression increased resistance to anticancer agents in human head and neck cell lines [29] . In NPC cell lines exposed to ionizing radiation, ZEB1 expression increased and inhibited AKT activation increased radiation sensitivity [25] . Forced overexpression of ZEB1 dramatically reduced the radiosensitivity induced by NEAT1 knockout and NEAT1 positively regulated ZEB1 expression through miR-204, which was in line with our results [30] . Moreover, inhibited AKT activation eliminated the radioresistance of CNE2 cells induced by overexpressed MINCR. AKT activation has been demonstrated to correlate with the metastasis, mortality, and radiotherapy resistance of NPC cells [31] . Interestingly, inhibition of ZIP4 inhibited metastasis, dissemination and invasion, and enhanced the radiosensitivity in human NPC cells by inactivating the PI3K/AKT axis [32] . miR-223 regulated the growth, invasion and chemotherapeutic resistance of glioblastoma stem cells to temozolomide via the PI3K/ AKT signaling pathway [33] .

In a word, we provided compelling evidences to state that lncRNA MINCR worked as a ceRNA for miR-223 to positively regulate ZEB1, and MINCR silencing strengthened the NPC cell radiosensitivity through the miR-223/ZEB1 axis and inactivating the PI3K/AKT axis. This study may offer new perspective for further understanding of NPC and finding new targets for effective therapies. Analysis of abovementioned results may give the chance to create the background for future clinical investigation and application in NPC. Thus, more studies in radiosensitivity of NPC cells are required in the future to develop clinical values.

QMZ is the guarantor of integrity of the entire study and contributed to the concepts and design of this study; YFC Figure 7 . MINCR knockdown reduces irradiation resistance of CNE2R cells in vivo. CNE2R cells stably MINCR-siRNA and scramble siRNA were inoculated subcutaneously into BALB/c nude mice at 5 × 10 6 per mouse (n = 3 in each group). Tumor growth was measured continuously every 5 days, and 20 days later, tumor growth was monitored every 3 days. Irradiation treatments were performed with 5 Gy of irradiation on day 15 and 25 after subcutaneously implantation. At 35 days post-implantation, the mice were euthanized by carbon dioxide asphyxiation. a. Relative tumor volume in CNE2R xenograft tumor models stably transfected with MINCR-siRNA; b. Relative tumor weight in CNE2R xenograft tumor models stably transfected with MINCR-siRNA; c, Relative content of p-AKT-and Ki67-positive tumor cells detected by immunohistochemical staining. In panels A and C, two-way ANOVA was used to determine statistical significance of quantification of immunostaining, whereas in panel B one-way ANOVA was used. *, p < 0.05, compared to the mock group. AKT, protein kinase B; ANOVA, analysis of variance.

contributed to the experimental studies and clinical studies; ZLC contributed to the data and statistical analysis; QMZ took charge of the manuscript preparation; ZLC contributed to the manuscript review. All authors read and approved the final manuscript.

No potential conflict of interest was reported by the authors.

Overexpression of osteopontin promotes cell proliferation and migration in human nasopharyngeal carcinoma and is associated with poor prognosis

miR-324-3p suppresses migration and invasion by targeting WNT2B in nasopharyngeal carcinoma

Cured nasopharyngeal carcinoma stage iv with haemostatic-intent radiotherapy

Genetic predisposition factors and nasopharyngeal carcinoma risk: a review of epidemiological association studies, 2000-2011: rosetta Stone for NPC: genetics, viral infection, and other environmental factors

MicroRNA-324-3p regulates nasopharyngeal carcinoma radioresistance by directly targeting WNT2B

The role of MALAT1/miR-1/ slug axis on radioresistance in nasopharyngeal carcinoma

MINCR is a MYC-induced lncRNA able to modulate MYC's transcriptional network in Burkitt lymphoma cells

Interaction and cross-talk between non-coding RNAs

Clinical significance of serum miR-223, miR-25 and miR-375 in patients with esophageal squamous cell carcinoma

miR-223 reverses the resistance of EGFR-TKIs through IGF1R/PI3K/Akt signaling pathway

Activation of miR-21 by STAT3 induces proliferation and suppresses apoptosis in nasopharyngeal carcinoma by targeting PTEN gene

MiR-1204 sensitizes nasopharyngeal carcinoma cells to paclitaxel both in vitro and in vivo

Circulating miR-17, miR-20a, miR-29c, and miR-223 combined as non-invasive biomarkers in nasopharyngeal carcinoma

Comparison of the 6th and 7th editions of the AJCC/UICC TNM staging system for gastric cancer focusing on the "N" parameter-related survival: the monoinstitutional NodUs Italian study

BMI1 and MEL18 promote colitis-associated cancer in mice via REG3B and STAT3

A broadly reactive one-step SYBR Green I real-time RT-PCR assay for rapid detection of murine norovirus

Binding of vascular anticoagulant alpha (VAC alpha) to planar phospholipid bilayers

Long non-coding RNA MINCR promotes gallbladder cancer progression through stimulating EZH2 expression

Cell-PLoc: a package of web servers for predicting subcellular localization of proteins in various organisms

starBase v2.0: decoding miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction networks from large-scale CLIP-Seq data

MicroRNA-223 enhances radiation sensitivity of U87MG cells in vitro and in vivo by targeting ataxia telangiectasia mutated

The EMT-activator Zeb1 is a key factor for cell plasticity and promotes metastasis in pancreatic cancer

RHBDD1 promotes colorectal cancer metastasis through the Wnt signaling pathway and its downstream target ZEB1

ATM-mediated stabilization of ZEB1 promotes DNA damage response and radioresistance through CHK1

Effect of AKT inhibition on epithelial-mesenchymal transition and ZEB1-potentiated radiotherapy in nasopharyngeal carcinoma

Systematical analysis of lncRNA-mRNA competing endogenous RNA network in breast cancer subtypes

LncRNA MINCR activates Wnt/beta-catenin signals to promote cell proliferation and migration in oral squamous cell carcinoma

MiR-223 targeting MAFB suppresses proliferation and migration of nasopharyngeal carcinoma cells

MiR-223-3p inhibits angiogenesis and promotes resistance to cetuximab in head and neck squamous cell carcinoma

The long non-coding RNA NEAT1 regulates epithelial to mesenchymal transition and radioresistance in through miR-204/ZEB1 axis in nasopharyngeal carcinoma

Activation of AKT is associated with metastasis of nasopharyngeal carcinoma

Inhibition of ZIP4 reverses epithelial-to-mesenchymal transition and enhances the radiosensitivity in human nasopharyngeal carcinoma cells

MiR-223/PAX6 axis regulates glioblastoma stem cell proliferation and the chemo resistance to TMZ via regulating PI3K/Akt pathway