MIR205HG acts as a ceRNA to expedite cell proliferation and progression in lung squamous cell carcinoma via targeting miR-299-3p/MAP3K2 axis

Human normal lung epithelial cell (BEAS-2B) and lung cancer cells (NCI-H520, SK-MES-1, NCI-H1703) were all bought from Chinese Academy of Sciences (Beijing, China). The above cells were inoculated in RPMI-1640 medium (Invitrogen, Carlsbad, CA, USA) adding 10% FBS (Invitrogen), then mixed with 1% penicillin/streptomycin (Sigma-Aldrich, Milan, Italy) and cultivated in a 5% CO₂ incubator at 37 °C. medium for incubation was required to change every 3 days.

Cells were put into a fresh 6-well plate until cells were passaged at 80% confluence. NCI-H520 and SK-MES-1 cells were co-transfected with shRNAs targeting MIR205HG (sh-MIR205HG#1/#2) and sh-NC. The pcDNA3.1/MIR205HG, pcDNA3.1/MAP 3 K2 and the empty pcDNA3.1 were constructed by Genechem (Shanghai China). The miR-299-3p mimics and NC-mimics were also gained from Genechem. All transfections were conducted by Lipofectamine 2000 (Invitrogen).

Total RNAs of cells were extracted by TRIzol reagent which was purchased from Invitrogen. The total RNAs were reverse-transcribed into cDNA under a Reverse Transcription Kit (Invitrogen). qRT-PCR analysis was achieved using SYBR Green Premix PCR Master Mix (Roche, Mannheim, Germany) under ABI HT9600 (Applied Biosystems, Foster City, CA, USA). The fold expression changes were calculated by using 2^-∆∆Ct method, and GAPDH/U6 was seen as the endogenous control.

SK-MES-1 and NCI-H520 cells were plated into 6-well plates and cultured after transfection. After 14 days, cells were cleaned with PBS (Sigma-Aldrich) and fixed in paraformaldehyde (Sigma-Aldrich) for 10 min, dyed by crystal violet (Sigma-Aldrich) for 5 min. Finally, the visible colony numbers were calculated manually.

Serum-free medium containing SK-MES-1 and NCI-H520 cells was placed in the top compartment and the lower compartment was filled with 10% FBS (Sigma-Aldrich). Matrigel (Corning Incorporated, NY, USA) was additionally pre-coated on the apical compartment for migration assay. After cultivating for 48 h, the bottom of the chamber was fixed by methanol (Sigma-Aldrich), and then dyed with crystal violet (Sigma-Aldrich). The cells that had migrated were counted at 5 randomly chosen visual fields using an inverted microscope (Olympus, Tokyo, Japan).

When reached to 80–90% confluent, cells were collected for total protein extraction by using RIPA cell lysis. After the concentration measured via a Pierce Bicinchoninic acid (BCA) Protein detection kit (Bio-Rad Laboratories, Hercules, CA, USA), proteins were isolated by performing SDS-PAGE and then moved onto a PVDF membrane (Beyotime, Shanghai, China). The membrane was sealed with 5% fat-free milk and rinsed by Tris-buffered saline/Tween 20 (TBST), followed by overnight incubation with the primary antibodies at 4 °C. The primary antibodies used in this study were as below: anti-E-cadherin (ab194982, Abcam), anti-N-cadherin (ab202030, Abcam), anti-Vimentin (ab193555, Abcam), anti-MAP 3 K2 (ab33918, Abcam) and antiGAPDH (ab8245, Abcam). Following washing in TBST, the membranes were subjected to 2 h of incubation with secondary antibodies at room temperature. The, the blots were visualized by the application of enhanced chemiluminescence (ECL; Amersham Pharmacia Biotec, Buckinghamshire, UK). GAPDH served as the internal control. The experiment was conducted in triplicate.

Cell apoptosis was examined at 48 h after transfection using FITC Annexin V/PI Apoptosis Detection Kit I (Ribobio, Guangzhou, China). In short, cells were reaped and cleaned in PBS (Sigma-Aldrich) and resuspended in 70% cooled ethanol (Sigma-Aldrich). The apoptosis ratio was determined by using FlowJo software (Ashland, OR, USA).

RIP experiments were performed using the Magna RNA-Binding Protein Immunoprecipitation Kit (Millipore, MA, USA). Then, cell lysates were cultivated with RIP buffer containing magnetic beads conjugated with human anti-Ago2 (ab32381, Abcam) or anti-IgG (Millipore, MA, USA). Furthermore, qRT-PCR was performed to evaluate the expression levels of MIR205HG, miR-299-3p and MAP 3 K2. Normal IgG was considered as control.

The wild-type and mutant binding sites of MIR205HG or MAP 3 K2 to miR-209-3p were sub-cloned into pmirGLO dual-luciferase vector to construct plasmids and then co-transfected with miR-299-3p mimics and pcDNA3.1/MIR205HG into NCI-H520 or SK-MES-1 cells, separately. Dual-luciferase reporter system was applied for detecting the activity of luciferase.

MicroRNA-205 host gene (MIR205HG) has been proofed to accelerate cell growth in HNSCC by down-regulating miR-590-3p [8]. Here in the first place, we applied GEPIA dataset to examine expression of MIR205HG in LUSC tissues. The results depicted that MIR205HG was significantly up-regulated in LUSC tissues compared with the adjacent normal tissues (Fig. 1a). Then, qRT-PCR detected that MIR205HG expression was relatively high in LUSC cell lines (NCI-H520, SK-MES-1 and NCI-H1703) compared with the normal control (Fig. 1b). Besides, NCI-H520 and SK-MES-1 cells were singled out for the forthcoming experiment for that they the higher level of MIR205HG.

Then, two shRNAs targeting MIR205HG were transfected into NCI-H520 and SK-MES-1 cells to detect the efficiency of down-regulating MIR205HG. Consequently, as depicted in Fig. 1c, the expression level of MIR205HG in selected cell lines were decreased conspicuously. Functionally, number of colonies declined distinctly in response to MIR205HG down-regulation (Fig. 1d). On the contrary, cell apoptosis ability was enhanced obviously, as demonstrated in Fig. 1e. In parallel, transwell assay showed that migration abilities of both cell lines were impaired observably (Fig. 1f). Moreover, from the results of western blot analysis, the expression of the epithelial marker (E-cadherin) was elevated observably while that of the mesenchymal markers (N-cadherin and Vimentin) was reduced comparably (Fig. 1g, Supplementary Figure 1A and Supplementary information file 1A). The above findings illustrated that MIR205HG had tumor-promoting function in LUSC.

We obtained potential miRNAs which had binding sites with MIR205HG via starBase database. Then through searching out the expression level of candidate miRNAs in LUSC from dbDEMC 2 website, 3 miRNAs (miR-299-3p, miR-143-3p and miR-122-5p) with low expression were selected out (Fig. 2a). RIP assay was applied to verify the binding situations between 3 miRNAs and MIR205HG. As denoted in Fig. 2b, only miR-299-3p and MIR205HG were significantly enriched in Ago2 precipitated RNA-induced silencing complexes (RISCs), indicating that miR-299-3p and MIR205HG involved in ceRNA axis. It further bolted miR-299-3p as our target miRNA. Figure 2c depicted the binding sites between miR-299-3p and wild type MIR205HG (MIR205HG-WT)/mutant MIR205HG (MIR205HG-Mut). Luciferase reporter assay exhibited that the luciferase activity of MIR205HG-WT obviously declined by elevated miR-299-3p overexpression, whereas that of MIR205HG-Mut bore little alteration (Fig. 2d). Besides, the expression of miR-299-3p was up-regulated dramatically under the condition of depleting MIR205HG in both cell lines (Fig. 2e). Hence, we concluded that miR-299-3p was the downstream molecule of MIR205HG in LUSC.

To forecast the possible target of miR-299-3p, we acquired the common target genes from 3 databases, including RNA22, miRmap and miRanda (Fig. 3a). There were seven candidate target genes of miR-299-3p (Fig. 3b). Previous studies have reported that MAP 3 K2 acted as an oncogene in the progression of certain types of cancer [18, 19]. Thus, MAP 3 K2 was chosen as the target gene of miR-299-3p. Simultaneously, miR-299-3p and MAP 3 K2 had binding sites, which were displayed in Fig. 3c. Luciferase reporter assay was used to further prove binding relationship between miR-299-3p and MAP 3 K2. As illustrated in Fig. 3d, miR-299-3p overexpression decreased the luciferase activity of MAP 3 K2-WT group via binding to MAP 3 K2-WT, while the luciferase activity of mutant MAP 3 K2 (MAP 3 K2-Mut) group had no alteration. Similarly, RIP assay showed that MIR205HG, miR-299-3p and MAP 3 K2 were all enriched in Ago2 containing beads in both cell lines (Fig. 3e). Via western blot analysis, the expression level of MAP 3 K2 was decreased with the down-regulation of MIR205HG, then increased with the overexpression of MAP 3 K2 (Fig. 3f, Supplementary Figure 1B and Supplementary information file 1B). To draw a conclusion, MAP 3 K2 acted as the target gene of miR-299-3p and indirectly regulated by MIR205HG.

We carried out rescue function assays to verify the effects of modulating relationship between MIR205HG and MAP 3 K2 on LUSC development. The results revealed that overexpressing MAP 3 K2 could restore number of colonies reduced by reducing MIR205HG (Fig. 4a). In contrast, the overexpression of MAP 3 K2 decreased cell apoptosis up-regulated by MIR205HG knockdown (Fig. 4b). Transwell assay also proved that elevating MAP 3 K2 could recover the reduced cell migration abilities caused by silenced MIR205HG (Fig. 4c). Evidence from western blot analysis suggested that overexpression of MAP 3 K2 decreased the expression of E-cadherin that was up-regulated by MIR205HG knockdown. N-cadherin and Vimentin expressions which were decreased by MIR205HG knockdown were enhanced by overexpression of MAP 3 K (Fig. 4d, Supplementary Figure 1C and Supplementary information file 1C). Altogether, MIR205HG could up-regulate MAP 3 K2 expression by sponging miR-299-3p in LUSC, thereby regulating LUSC cell proliferation, apoptosis, migration and EMT process.