LncRNA MIR205HG regulates melanomagenesis via the miR-299-3p/VEGFA axis

Knockdown of MIR205HG significantly reduces proliferation of melanoma cells

Quantitative real-time PCR (qRT-PCR) analysis showed that MIR205HG levels were significantly upregulated in melanoma cancer cells (A375, MNT-1 and SK-MEL-28 cells) compared to the normal human melanocytes (Figure 1A). Next, qRT-PCR analysis showed that MIR205HG levels were significantly reduced in the MIR205HG-shRNA1 and MIR205HG-shRNA2-transfected A375 and MNT-1 cells compared to their corresponding controls (Figure 1B, 1C). MIR205HG knockdown was higher in the melanoma cells with MIR205HG-shRNA2 compared to MIR205HG-shRNA1. Hence, we selected MIR205HG shRNA2 for further experiments. CCK-8 assay results showed that knockdown of MIR205HG in A375 and MNT-1 cells significantly reduced cell viability compared to the corresponding controls (Figure 1D, 1E). Furthermore, TCGA database analysis showed that high expression of MIR205HG was closely associated with lower survival rate of patients with melanoma (Figure 1F). Besides, QRT-PCR analysis showed that MIR205HG was significantly upregulated in melanoma tissues compared to the adjacent normal skin tissues (n=30; Figure 1G). Taken together, these results demonstrate that knockdown of MIR205HG significantly suppressed proliferation of melanoma cells.

MIR205HG directly binds to miR-299-3p in melanoma cells

We explored the miRDB (http://www.mirdb.org/) and starBase (https://web.archive.org/web/20110222111721/http://starbase.sysu.edu.cn/) database, and then identified miR-299-3p as the most potential target miRNA of MIR205HG (Figure 2A). In addition, miR-299-3p has been reported to play a key role in progression of malignant tumor [20]. Therefore, miR-299-3p was selected from starBase and miRDB. Furthermore, dual luciferase reporter assay results confirmed that miR-299-3p was the downstream target miRNA of MIR205HG (Figure 2B). QRT-PCR analysis showed that miR-299-3p levels were significantly higher in A375 cells transfected with miR-299-3p mimics and significantly reduced in those transfected with the miR-299-3p inhibitor (Figure 2C). RNA pull-down assay showed significant enrichment of miR-299-3p with the biotinylated MIR205HG probe compared to the corresponding control (Figure 2D). Overall, these results confirmed that miR-299-3p was directly bound by MIR205HG.

Silencing of MIR205HG induces apoptosis of melanoma cells

Flow cytometry analysis showed that knockdown of MIR205HG significantly increased apoptosis in the A375 and MNT-1 cells compared to the corresponding controls (Figure 3A–3D). However, treatment with the miR-299-3p inhibitor reduced apoptotic rates in the MIR205HG-knockdown A375 and MNT-1 cells (Figure 3A–3D). This suggested that MIR205HG silencing increased apoptosis of melanoma cells via miR-299-3p. We chose MIR205HG-shRNA2-transfected A375 cells for further experiments because they were more sensitive than the MIR205HG-shRNA2-transfected MNT-1 cells.

MIR205HG silencing inhibits migration and invasion of melanoma cells via miR-299-3p

Next, we analyzed the effects of MIR205HG knockdown on the invasiveness and migration of A375 melanoma cells using the transwell and wound healing assays, respectively. MIR205HG knockdown significantly reduced the invasiveness of A375 cells, but these effects were partially rescued by treatment with the miR-299-3p inhibitor (Figure 4A, 4B). Furthermore, MIR205HG knockdown significantly reduced the migration of A375 cells, but, these effects were significantly reversed by treatment with the miR-299-3p inhibitor (Figure 4C, 4D). Taken together, these data demonstrate that MIR205HG silencing inhibits migration and invasion of melanoma cells via miR-299-3p.

MiR-299-3p directly binds to 3’UTR of VEGFA in melanoma cells

We analyzed the targetscan (http://www.targetscan.org/vert_71/) and miRDB (http://www.mirdb.org/) database, and then identified 3’-UTR of VEGFA as a potential target of miR-299-3p (Figure 5A). Since VEGFA is involved in tumorigenesis of melanoma [21, 22], it was selected for further analysis. Furthermore, we confirmed that VEGFA was a direct target of miR-299-3p using the dual luciferase report assay (Figure 5B). In addition, miR-299-3p mimics significantly inhibited the expression of VEGFA in melanoma cells in a dose-dependent manner (Figure 5C). QRT-PCR analysis showed that the expression of VEGFA mRNA was significantly downregulated in MIR205HG shRNA2-transfected A375 cells compared to the corresponding controls, but VEGFA mRNA expression was restored partially by treatment with the miR-299-3p inhibitor (Figure 5D). Western blot analysis showed that cleaved caspase-3 and E-cadherin levels were significantly upregulated in MIR205HG-silenced A375 cells compared to the corresponding controls, but were significantly reduced in the MIR205HG-silenced A375 cells treated with the miR-299-3p inhibitor (Figure 5E, 5F). In contrast, MIR205HG shRNA2-transfected A375 cells showed reduced levels of VEGFA, α-SMA and Vimentin proteins, but these effects were reversed by treatment with the miR-299-3p inhibitor (Figure 5E, 5F). These results demonstrate that MIR205HG regulates melanoma progression via the miR-299-3p/VEGFA axis.

Downregulation of MIR205HG or miR-299-3p agomir significantly suppressed the growth of melanoma in vivo

Finally, we analyzed the role of MIR205HG in melanoma by generating the xenograft melanoma model mice. We observed that tumor volumes (Figure 6A), tumor sizes (Figure 6B), and tumor weights (Figure 6C) were significantly reduced in nude mice xenografted with MIR205HG-silenced A375 cells compared to the controls. Moreover, VEGFA protein levels were significantly downregulated (Figure 6D, 6E) and cleaved caspase-3 levels were significantly increased (Figure 6D, 6E) in tumor tissues derived from MIR205HG-silenced A375 cells compared to the controls. Additionally, the data of IHC staining indicated that knockdown of MIR205HG notably inhibited the expression of VEGFA but increased the level of active caspase 3 in tumor tissues of mice (Figure 6F). On the other hand, miR-299-3p agomir significantly decreased the tumor size and weight in mice, while downregulation of miR-299-3p notably promoted the tumor growth of melanoma in mice (Supplementary Figure 1A–1C). Taken together, these results demonstrate that MIR205HG silencing or miR-299-3p agomir significantly inhibits the in vivo growth of melanoma.

Melanoma cell lines and cell culture

We purchased human melanocytes, melanoma cell lines (MNT-1, A375, and SK-MEL-28), and 293T cell lines from the American Type Culture Collection (ATCC, Manassas, VA, USA). They were cultured in RPMI-1640 medium (ThermoFisher, Shanghai, China) with 10% fetal bovine serum (FBS) in a humidified incubator maintained at 37° C and 5% CO₂.

Cell transfections

We obtained lentiviral vector (pLVX-IRES-Puro) cloned with the negative control (NC) or short-hairpin RNAs against lncRNA MIR205HG (MIR205HG shRNA1 or MIR205HG shRNA2) from the Hanbio Biotechnology Co., Ltd (Shanghai, China). The negative control (NC) and MIR205HG shRNA1 or MIR205HG shRNA2 lentiviral vectors were transfected into 293T cells and the cells were incubated at 37°C for 48 h. Subsequently, we removed cell debris and cells by centrifugation at 956×g for 15 min and passed the supernatants through a 45 μm filter (Costar, Cambridge, MA, USA) to purify the lentiviral particles. We then centrifuged melanoma cells (5×10⁶/well) with the purified lentiviruses at 956×g for 15 min and then incubated the cells for 48 h in RPMI-1640 medium containing puromycin (Sigma, MA, USA) for selection. The efficiency of MIR205HG knockdown was verified by qRT-PCR.

We purchased miR-299-3p mimics, miR-299-3p inhibitor or negative control RNA (NC) from GenePharma (Shanghai, China) and transfected them into melanoma cells using Lipofectamine 2000 as previously described [45]. The efficiency of transfection was verified by qRT-PCR.

Melanoma patient tissues

We collected 30 pairs of melanoma and adjacent normal tissues between December 2018 and December 2019 from melanoma patients at the Xinxiang Central Hospital. We also obtained clinical and pathological data of these patients as well as their written informed consent. The tissue samples were stored at -80° C. The present study was approved by the Ethics Committee of the Xinxiang Central Hospital. The information of patients has been supplemented in Table 1.

Quantitative real time polymerase chain reaction (qRT-PCR)

Total RNA was extracted from the melanoma cell lines using TRIzol reagent (TaKaRa, Tokyo, Japan) according to the manufacturer's protocol. First-strand cDNA synthesis was performed using the PrimeScript RT reagent Kit (Takara) according to the manufacturer's protocol. Then, quantitative PCR (qPCR) was performed in an ABI7500 real-time PCR system using SYBR green(Takara). The qRT-PCR protocol was 94° C for 2 mins followed by 35 cycles of 94° C for 30 s and 55° C for 45 s.

All the qRT-PCR primers were obtained from GenePharma (Shanghai, China). These include

MIR205HG (forward): 5’-GACCGTTGTTAGCACGCCTT-3’;

MIR205HG (reverse): 5’-CACGTATCGGTCCGTGTTGG-3’;

miR-299-3p (forward): 5’-TTCCATACTGCAACGCCATACC-3’;

miR-299-3p (reverse): 5’-GCAATCCGCCCTTAGTCCAA-3’;

VEGFA (forward): 5’-GAACTTTCTGCTGTCTTGGGTG-3’;

VEGFA (reverse): 5’-GGCAGTAGCTGCGCTGATAG-3’;

U6 (forward): 5’-CGTCTTCCCAGGACCGTA-3’;

U6 (reverse): 5’-CGAATCCTGACATTAAGTCG-3’;

β-actin (forward): 5’-CCTGCGAAACACCTTGATCG-3’, and

β-actin (reverse): 5’-TCGTCATGTTCCCCACTTCG-3’.

U6 was used as reference to quantify miR-299-3p levels. β-actin was used to quantify MIR205HG levels. The relative levels of miR-299-3p and MIR205HG expression were evaluated using the 2^-ΔΔCT method.

The cancer genome atlas (TCGA)

The correlation between MIR205HG expression and survival rate of patients with melanoma was analyzed by TCGA. The analysis included the data from 531 patients with melanoma. Among the patients with melanoma, 209 patients had high expression of MIR205HG, while the others had low level of MIR205HG. Meanwhile, the data of TCGA was analyzed from Gene Expression Profiling Interactive Analysis (GEPIA) as previously described [46].

CCK-8 assay

Cell counting kit-8 (CCK8, Beyotime, Shanghai, China) assay was used to analyze cell viability. We seeded 5×10³ MNT-1 or A375 cells per well in the blank, negative control (NC) or MIR205HG knockdown (MIR205HG shRNA2) groups in 96-well plates for 0, 24, 48 and 72 h, respectively. Subsequently, we added 10 μl CCK-8 reagent and incubated cells further for 2 h at 37° C. Then, we measured absorbance at 450 nm using a microplate reader (Thermo Fisher Scientific, Waltham, MA, USA).

RNA pull-down

We used the Biotin RNA Labeling Mix (Roche, Basel, Switzerland) according to the manufacturer’s instructions to generate the biotinylated control and MIR205HG probes. We then transfected biotinylated-MIR205HG or control probes into the A375 cells. The RNA structure buffer (Thermo Fisher Scientific, MA, USA) was used to induce secondary structure formation in the biotin-labeled MIR205HG or control RNAs. The streptavidin beads (Thermo Fisher Scientific, Waltham, MA, USA) were washed three times with the 500 μL of RNA immunoprecipitation wash buffer (Thermo) and then incubated with the biotinylated RNAs at 4° C overnight. The overnight-incubated mixture was separated by a magnetic field to obtain streptavidin bead-RNA complexes. Then, the A375 cell lysates were incubated with the streptavidin bead-RNA complexes on a rotator at room temperature for 1 h followed by separation with a magnetic field to obtain streptavidin bead-RNA-protein complexes, which were then analyzed by qRT-PCR and western blotting.

Western blotting

Total protein lysates from the melanoma cells were prepared by incubation with the RIPA buffer (Cell Signaling Technology, Danvers, MA, USA) on ice according to the manufacturer’s instructions. Then, equal amounts of total protein lysates were separated on a 10% SDS-PAGE. The separated proteins were transferred onto the polyvinylidene difluoride (PVDF) membranes. The membranes were then blocked with 5% BSA (Gibco, Grand Island, NY, USA) in Tris-buffered saline (TBS) containing 0.5% Tween-20 (TBST) for 60 min. The blots were then incubated overnight at 4° C with primary antibodies against VEGFA (1:1000, Abcam, CA, USA), cleaved caspase-3 (1:1000, Abcam), α-SMA (Abcam, 1:1000), E-cadherin (Abcam, 1:1000), vimentin (Abcam, 1:1000), and β-actin (Abcam, 1:1000). Then, after washing three times with 1X TBST buffer for 5 min, the membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG polyclonal secondary antibody (1:5000; Beyotime Biotechnology, Shanghai, China) at room temperature for 1 h. The blots were developed with the ECL+Plus chemoluminescence western blot system kit (Amersham, Cytiva, Shanghai, China). The density of the protein bands was measured using the ImageJ software.

Cell apoptosis

We centrifuged 1×10⁶ MNT-1 or A375 cells per well in a 6-well plate at 1000 rpm for 5 min. Then, after removing the supernatant, we incubated the cells with 5 μl Annexin V-FITC (20 μg/ml) and 5 μl propidium (PI; 50 μg/ml) in 100 μl Annexin-V binding buffer for 15 min in the dark. The stained cells were then analyzed in a BD flow-cytometer (BD, Franklin Lake, NJ, USA) and the proportions of apoptotic cells (Annexin-V⁺ PI^- plus Annexin-V⁺ PI⁺) were estimated using the Fluorescence activated Cell Sorting (FACS, BD, Franklin Lake, NJ, USA) and Flowjo software (BD, Franklin Lake, NJ, USA).

Dual luciferase reporter assay

The wild-type and mutant constructs of MIR205HG were cloned into the pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, USA). Then, melanoma cells were co-transfected with wild-type or mutant VEGFA 3’UTR and miR-NC or miR-299-3p mimics using Lipofectamine 3000. The transfection efficacy was analyzed using pmirGLO reporter as an internal control. The luciferase activities were analyzed at 48 h using the Dual-Luciferase Reporter Assay System (Promega, USA) according to the manufacturer’s instructions. The parts, sequences and transcripts of MIR205HG were presented at https://www.ncbi.nlm.nih.gov/nuccore/NR_145433.1. The information originated from National Center for Biotechnology Information (NCBI).

Transwell invasion assay

For the in vitro invasion assay, the upper chambers of the Transwell plates (Corning, New York, NY, USA) were coated with 100 μl of Matrigel (BD Biosciences, Franklin Lake, NJ, USA). Then, we seeded the melanoma cells (1×10⁵ cells) in FBS-free RPMI-1640 medium into the upper chamber, and added RPMI-1640 supplemented with 10% FBS into the lower chamber. The cells were incubated at 37° C for 24 h in a humidified incubator maintained at 5% CO₂. Then, the cells attached to the underside of the membrane were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet solution. Finally, we captured images and counted the numbers of invading cells under a light microscope.

Wound healing assay

We seeded 5×10³ A375 cells into each well of a 24-well cell culture cluster plate. After the cells reached 80-90% confluence, we scraped the cell monolayer with a small pipette head. Then, after washing thrice with PBS, we added serum-free medium and cultured the cells further. We recorded the scratch widths at 0 and 48 h using an optical microscope. The experiment was repeated 3 times.

Animal study

The in vivo animal experiments were performed in accordance with the National Institutes of Health (NIH) guide for the care and use of laboratory animals and the protocol was approved by the Ethics Committees of Xinxiang Central Hospital. We purchased eight 6-week old BALB/nude mice from Vital River (Beijing, China). The control, MIR205HG shRNA2, miR-299-3p agomir or miR-299-3p antagomir-transfected A375 cells (1×10⁶) were subcutaneously injected into the right flanks of the nude mice (4 mice per group) as previously described [47]. Tumor growth was measured weekly in all mice for four weeks. Tumor volume was calculated as length× (width)²/2 as previously described [8]. The animals were sacrificed at 4 weeks after xenografting the melanoma cells. The tumor tissues were harvested and weighed, and the expression of VEGFA and active caspase 3 in tumor tissues of mice were detected by immunohistochemistry (IHC) staining as previously reported [48].

Statistical analysis

All experiments were performed at least thrice independently. The data are represented as means ± standard deviation (SD). The data between two groups were compared using the unpaired Student’s t-test, whereas, data between multiple groups were compared using one-way analysis of variance (ANOVA) followed by Tukey test (GraphPad Prism7). P<0.05 was considered statistically significant.