The human tongue cancer cell lines, CAL-27, SCC-9, SCC-4, SCC-15 and SCC-25 were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). Cells from a mixture of three tissues were used as the control.

Thirty paired tongue cancer specimens and adjacent non-neoplastic tongue tissues were collected from patients following tumor surgical resection at The Affiliated Zhongshan Hospital, Sun Yat-sen University (Zhongshan, China). All the human tissues were obtained with informed consent from all subjects and donors. This study was approved by the Clinical Research Ethics Committee of Zhongshan Hospital.

miRNA mimics and siRNAs were synthesized by GenePharma Co., Ltd. (Shanghai, China). Transfection was conducted using Lipofectamine™ 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA), according to the protocol recommended by the manufacturer. After a 48-h transfection, the cells were collected and used for further experiments.

Cell proliferation was determined using an MTT assay (Promega Corp., Madison, WI, USA), according to the manufacturer's instructions. Twenty-four hours after being seeded into 96-well plates at a density of 5,000 cells/well, the cells were transfected with 100 nM miR-124 mimics/miR-124NC or si-NC/si-MALAT1 or si-NC/si-JAG1. Twenty-four hours after transfection, 20 µl of 5 mg/ml MTT was added and then the cells were incubated for 4 h in a humidified incubator. DMSO (200 µl) was added to dissolve the formazan after the supernatant was discarded. The optical density (OD) was measured at 490 nm.

Total RNA was extracted by TRIzol reagent (Invitrogen) following the manufacturer's instructions. Then a High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA) was used to reversely transcribe RNA samples. Quantitative RT-PCR was performed using the FastStart Universal SYBR-Green Master (Roche, Indianapolis, IN, USA). The primers were as follows: MALAT1 sense, 5′-AAAGCAAGGTCTCCCCACAAG-3′ and antisense, 5-GGTCTGTGCTAGATCAAAAGGCA-3′; GAPDH sense, 5′-AGAAGGCTGGGGCTCATTTG-3′ and antisense, 5′-AGGGGCCATCCACAGTCTTC-3′. The relative fold change of candidate genes was analyzed using the 2^−ΔΔCT method.

After removal of cells by trypsinization, a 1×10⁵ cell suspension with 10% FBS was added into the Transwell inserts with an 8-µm pore size, in 24-well plates coated with 50 µl Matrigel (both from BD Biosciences). Then 200 µl of medium supplemented with 15% FBS was added into the bottom chamber; 24 and 48 h after migration, the cells which did not undergo migration in the upper chamber were wiped up; the filters were independently fixed with 4% paraformaldehyde and were stained with hematoxylin and eosin. Then the number of cells from five random fields was counted.

RIPA buffer (Sigma-Aldrich, St. Louis, MO, USA) was used to lyse cell lysates with a complete protease inhibitor cocktail (Roche). Cell lysates were transferred to a 1.5-ml tube and kept at −20°C before use. SDS-PAGE was conducted to separate the cellular proteins. All the cellular proteins in this study were separated by 5% stacking gel and 10% running gel. The candidate proteins whose molecular weights were included in the information for the Pre-Stained SeeBlue rainbow markers (Invitrogen) were loaded in parallel. The following antibodies were used to probe the membranes: JAG1 (Abcam, Cambridge, MA, USA), and β-actin (Sigma-Aldrich). The blots were detected on a Kodak film developer (Fujifilm, Tokyo, Japan).

With the use of Lipofectamine™ 2000 transfection reagent, CAL-27 cells cultured in 24-well plates were co-transfected with luciferase reporter plasmids and miRNA mimics as well as the internal control pRSV-β-galactosidase vector. Forty-eight hours post-transfection, CAL-27 cells were lysed with lysis buffer (25 mM Tris-phosphate, 1% Triton X-100, 1 mM DTT, 2 mM EDTA, 10% glycerol, pH 27.8). After centrifugation at 14,000 rpm for 3 min, the supernatant was transferred to a new 1.5-ml tube. The luciferase activity was assessed using the GloMax 20/20 Luminometer (Promega Corp.) after mixing 50 µl supernatant with 50 µl luciferase assay buffer (265 µM ATP, 2.70 mM MgSO₄, 1.07 mM MgCl₂, 135 µM coenzyme A, 20 mM Tricine, 0.1 mM EDTA, 33.3 mM DTT, 235 µM D-luciferin). The β-galactosidase activity from the pRSV-β-galactosidase vector was used for the normalization of the luminescence levels. O-nitrophenyl-β-galactoside (ONPG) colorimetric assays were performed to assess the β-galactosidase activity. The β-galactosidase activity was evaluated using the measurement of o-nitrophenol obtained on an ELISA plate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at the wavelength of 490 nm.

Experimental results are presented as the mean ± SD of at least three independent experiments. Comparisons between two groups were conducted using a two-tailed Student's t-test and differences were considered to be statistically significant at a P-value <0.05.

Initially the expression levels of MALAT1 in 30 paired samples (tongue cancer specimens and corresponding adjacent non-tumor tissues) were examined using real-time PCR. Results showed that the expression level of MALAT1 was markedly higher in tumor tissues, compared with the level in the control group (adjacent non-neoplastic tongue tissues) (Fig. 1A). Then MALAT1 expression levels in five human tongue cancer cell lines, CAL-27, SCC-9, SCC-4, SCC-15, SCC-25 and cells from a mixture of three tissues that were used as the control were determined using real-time PCR. The results revealed that in all human tongue cancer cell lines, the expression level of MALAT1 was upregulated compared to the control group (Fig. 1B).

Next we investigated the association of MALAT1 expression with tongue cancer cell proliferation and invasion. Knockdown of MALAT1 was accomplished using si-MALAT1 and the inhibitory efficiency was verified by real-time PCR (Fig. 2A). Two tongue cancer cell lines, CAL-27 and SCC-9, were transfected with si-NC or si-MALAT1, and then the cell proliferation and invasion were determined by MTT and Transwell assays. MTT assays revealed that knockdown of MALAT1 significantly attenuated the proliferation of both CAL-27 and SCC-9 cell lines over time, compared with the si-NC group (Fig. 2B). Transwell assays revealed that knockdown of MALAT1 markedly decreased the cell migratory abilities of both CAL-27 and SCC-9 cell lines, and that the percentage of invasion in the si-MALAT1 group was significantly decreased compared with the si-NC group (Fig. 2C). Collectively, these data revealed that lncRNA-MALAT1 was specifically upregulated in tongue cancer tissues and cell lines and promoted tongue cancer cell growth and invasion.

According to previous studies, miR-124 plays a suppressive role in cancer (15,16). To further investigate the mechanism by which MALAT1 regulates tongue cancer cell growth, we ascertained the association of miR-124 and MALAT1. The results from real-time PCR revealed that the expression levels of miR-124 in the CAL-27 and SCC-9 cell lines were significantly upregulated after knockdown of MALAT1 compared with expression levels of miR-124 in the the si-NC groups (Fig. 3A). Then miR-124 mimics and miR-124 inhibitor were used to achieve miR-124 overexpression and miR-124 inhibition, and the expression levels of miR-124 were assessed using real-time PCR in CAL-27 and SCC-9 cell lines (Fig. 3B). The expression levels of MALAT1 were determined using real-time PCR in response to miR-124 overexpression or miR-124 inhibition. Results showed that MALAT1 expression was decreased in response to miR-124 overexpression while MALAT1 expression was increased in response to miR-124 inhibition, compared with the miR-124 NC groups (Fig. 3C). Moreover, the expression levels of miR-124 were revealed to be significantly downregulated in tumor tissues compared with the adjacent non-neoplastic tongue tissues (Fig. 3D). We examined the potential correlation between the RNA expression levels of MALAT1 and miR-124 and an inverse correlation between their expression levels was observed (Fig. 3E). Collectively, miR-124 expression was correlated with MALAT1 and an inverse correlation between the expression of MALAT1 and miR-124 was observed.

In order to investigate the mechanism by which miR-124 is correlated with MALAT1, we constructed a wt-MALAT1 3′-untranslated region luciferase reporter vector (wt-MALAT1), and a mut-MALAT1 3′-untranslated region luciferase reporter vector (mut-MALAT1) by sequentially mutating the predicted two miR-124 binding sites in the MALAT1 3′-untranslated region (Fig. 4A). The wt-MALAT1/mut-MALAT1 vectors were co-transfected with miR-124 NC/miR-124 mimics into CAL-27 cells. The luciferase activity of the MALAT1 3′-untranslated region of the luciferase reporter vector was markedly attenuated in the cells co-transfected with the wt-MALAT1 and miR-124 mimics, compared to the control groups (Fig. 4B). In addition, suppression of MALAT1 3′-untranslated region luciferase reporter activity was eliminated in the cells co-transfected with mut-MALAT1 and miR-124 mimics (Fig. 4B).

It has been indicated that miR-124 functions as a tumor suppressor in several types of cancer (17,18). Here, an MTT assay was used to determine tongue cancer cell growth in response to miR-124 overexpression. It was observed that the cell proliferation of both CAL-27 and SCC-9 cell lines was decreased when miR-124 was overexpressed compared with the miR-124 NC group (Fig. 5A). The results from the Transwell assays indicated that miR-124 overexpression markedly decreased the cell migratory abilities of both CAL-27 and SCC-9 cell lines compared with the miR-124 NC group (Fig. 5B).

In a previous study, JAG1 was reported to be influenced by miR-124 (19). In order to reveal the association of miR-124 with JAG1 in tongue cancer, we created a wt-JAG1 3′-untranslated region luciferase reporter vector (wt-JAG1) and a mut-JAG1 3′-untranslated region luciferase reporter vector (mut-JAG1) by sequentially mutating the predicted two miR-124 binding sites in the JAG1 3′-untranslated region (Fig. 6A). The wt-JAG1/mut-JAG1 vectors with the miR-124 NC/miR-124 mimics were co-transfected into CAL-27 cells. The luciferase activity of the JAG1 3′-untranslated region luciferase reporter vector was markedly attenuated in the cells co-transfected with the miR-124 mimics and wt-JAG1, compared to the control groups (Fig. 6A). Additionally, miR-124-mediated suppression of the JAG1 3′-untranslated region luciferase reporter activity was eliminated in the cells co-transfected with miR-124 mimics and mut-JAG1 (Fig. 6A). Next, through western blot analysis assay we revealed that the expression of JAG1 was significantly downregulated by miR-124 overexpression but upregulated by miR-124 inhibition in both CAL-27 and SCC-9 cell lines (Fig. 6B). Moreover, the expression levels of JAG1 were revealed to be significantly upregulated in tumor tissues compared with levels in the adjacent normal tissues (Fig. 6C). We then examined the potential correlation between the RNA expression levels of JAG1 and miR-124 and observed an inverse correlation between the expression of JAG1 and miR-124 (Fig. 6D).

To further investigate the correlation between MALAT1 and JAG1, we determined the potential correlation between the RNA expression levels of MALAT1 and JAG1. A positive correlation between the expression levels of MALAT1 and JAG1 was observed (Fig. 7A). Western blot analysis results revealed that the protein contents of JAG1 in the CAL-27 and SCC-9 cell lines were significantly decreased in response to MALAT1-knockdown by si-MALAT1 or JAG1 inhibition by si-JAG1 (Fig. 7B and C).

Given that JAG1 was positively correlated with MALAT1, we next determined the role of MALAT1 and JAG1 in the regulation of tongue cancer proliferation and invasion. Transwell results revealed that the cell migratory abilities were decreased after JAG1 inhibition (Fig. 8A). Moreover, according to the MTT assay results, CAL-27 and SCC-9 cell growth was attenuated in response to JAG1 inhibition by si-JAG1 (Fig. 8B).