The present study was approved by the Krasnoyarsk State Medical University Local Ethics Committee (protocol no. 70/2016, issued on June 6, 2016, Krasnoyarsk, Russia).

Samples from primary melanomas (n=12; 5 males and 7 females; median age, 56 years; age range, 32–81 years) and benign melanocytic tumors (n=9, 4 males and 5 females; median age, 36 years; age range 14–65 years) were obtained from patients from the Krasnoyarsk Regional Oncology Center (Krasnoyarsk, Russia). Samples were taken during the period from June 2016 to October 2016. Skin biopsies were obtained after getting written informed consent. The tissues were resected and immediately immersed in 1xRNAlater^® stabilization solution (cat. no. AM7020; Ambion; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at room temperature for 1 to 2 h for transportation. Subsequently, all samples were stored at −20°C prior to use. The diagnosis was confirmed by certified pathologists using a light microscope (Leica DM2500; Leica Microsystems, Germany) of the Krasnoyarsk Pathological Anatomy Bureau. Melanoma was staged in accordance with the American Joint Committee on Cancer 2009 guidelines (14).

The human melanoma BRO cell line was obtained from the Research Institute of Fundamental and Clinical Immunology (Novosibirsk, Russia). The melanoma SK-MEL1 cell line (ATCC^® HTB-67™) was obtained from the A.N. Sysin Research Institute of Human Ecology and Environmental Health at the Ministry of Health of the Russian Federation (Moscow, Russia). The cells were cultured in RPMI-1640 with L-glutamine (Gibco; Thermo Fisher Scientific, Inc.) and 10% fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) at 37°C in 5% CO₂ in a CO₂-incubator (Sanyo MSO-5AC; Sanyo Electric Co., Ltd., Osaka, Japan).

For total RNA isolation, tissue samples were homogenized with liquid nitrogen and then suspended in 200 µl digestion buffer of the RecoverAll^™ Total Nucleic Acid Isolation kit (cat. no. AM1975; Ambion; Thermo Fisher Scientific, Inc.). RNA was isolated according to the manufacturer's protocol, and then eluted with 50 µl nuclease-free water. miRNA concentration was estimated by a Qubit^® 2.0 fluorimeter (Invitrogen; Thermo Fisher Scientific, Inc.) with the use of Qubit^® microRNA Assay kit (ref. Q32880; Thermo Fisher Scientific, Inc.).

RT-qPCR was performed using the TaqMan^™ miRNA Assay (cat. no. 4427975; Applied Biosystems; Thermo Fisher Scientific, Inc.). The reaction mixture for RT, which consisted of 4 µl RNA from each sample, 0.4 µl 5X primers from the TaqMan^™ miRNA assay, 2 µl 50X OT buffer, 1 µl DTT, 1 µl dNTPs mix and 0.5 µl revertase from the Moloney Murine Leukemia Virus (MMLV) RT kit (cat. no. SK021, Evrogen, Moscow, Russia), was incubated at 37°C for 30 min. Subsequently, 2 µl cDNA was added to the PCR cocktail containing 8 µl 2.5-fold RT-PCR reaction mixture with ROX reference dye (Syntol, Moscow, Russia), 1 µl 20X primers/probe mix from the TaqMan^™ miRNA Assay and 9 µl de-ionized water. U6 small nuclear RNA and RNU6B were used as endogenous controls (Applied Biosystems; Thermo Fisher Scientific, Inc.). The reaction was performed on a Step One^™ Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) with the following temperature cycling protocol: Preheating at 50°C for 2 min and 95°C for 10 min, followed by 40 cycles: Denaturation step at 95°C for 15 sec and annealing and elongations step at 60°C for 1 min with 6-carboxyfluorescein/ROX detection. Data were analyzed using the ΔΔCq method, as previously described (15).

To estimate the expression levels of target genes, RT of total RNA was performed using the MMLV RT kit (Evrogen) in a mixture containing 4 µl RNA sample, 0.5 µl random primers, 2 µl 50X OT buffer, 1 µl DTT, 1 µl dNTPs mix and 0.5 µl revertase. The 2 µl cDNA was used for PCR with primers specific to the genes under investigation (cat. no. 4331182, Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the aforementioned cycling protocol for the miRNA expression assay.

The BRO and SK-MEL1 cells were transfected 24 h prior to starting the subsequent experiments with specific anti-miR^™ miRNA inhibitors for miR-204-5p (mature miRNA sequence: UUCCCUUUGUCAUCCUAUGCCU) and miR-3065-5p (mature miRNA sequence: UCAACAAAAUCACUGAUGCUGGA) (both 5 nmol lyophilized pellet, Assay IDs, AM11116 and AM18328, respectively; cat. no. AM17000; Ambion; Thermo Fisher Scientific, Inc.) and with mirVana^® miRNA mimics in an additional experimental series (5 nmol lyophilized pellet, Assay IDs, MC11116 and MC18328, respectively; cat. no. 4464066, Ambion; Thermo Fisher Scientific, Inc.). When cells reached a final concentration of 1–6×10⁵ cells/ml, transfection experiments were performed using 1.5 µl Lipofectamine^® RNAiMAX Transfection Reagent (cat. no. 13778150; Invitrogen; Thermo Fisher Scientific, Inc., USA)/500 µl cells. Anti-miR or mimic solutions were added to the cells containing culture medium RPMI-1640 with L-glutamine (Gibco; Thermo Fisher Scientific, Inc.) to a final concentration of 25 nM. The medium was replaced every 24 h after transfection. To estimate the gain- or loss-of-function effect of microRNA, anti-miR^™ miRNA Inhibitor Negative Control #1 (5 nmol lyophilized pellet, cat. no. AM17010, Ambion; Thermo Fisher Scientific, Inc.) and mirVana^™ miRNA Mimic Negative Control #1 (5 nmol lyophilized pellet, cat. no. 4464058, Ambion; Thermo Fisher Scientific, Inc.) negative controls were used.

To evaluate the transfection efficiency, the effects of anti-miR™ hsa-let-7c miRNA inhibitor (cat. no. 4392431; Ambion; Thermo Fisher Scientific, Inc.) and hsa-miR-1 mimic (cat. no. 4464062; Ambion; Thermo Fisher Scientific, Inc.) positive controls were measured and compared with those of the respective negative controls. For this purpose, the cells were transfected using 1.5 µl Lipofectamine^® RNAiMAX Transfection Reagent (cat. no. 13778150; Invitrogen; Thermo Fisher Scientific, Inc.) with anti-miR or mimic positive controls; anti-miR™ hsa-let-7c miRNA Inhibitor Positive Control (5 nmol lyophilized pellet, cat. no. 4392431, Ambion; Thermo Fisher Scientific, Inc.) and mirVana™ miRNA Mimic miR-1 Positive Control (5 nmol lyophilized pellet, cat. no. 4464063, Ambion; Thermo Fisher Scientific, Inc.) and incubated at 37°C with 5% CO₂ for 48 h. Subsequently, RNA was isolated from the cells and the changes in levels of the High-mobility group AT-hook 2. (HMGA2, Taqman^® Gene Expression Assays, cat. no 4331182, Applied Biosystems™; Thermo Fisher Scientific, Inc.) for the inhibitors and Twinfilin actin-binding protein 1 (TWF1, Taqman^® Gene Expression Assays, cat. no. 4331182, Applied Biosystems™; Thermo Fisher Scientific, Inc.) for the mimics were estimated by PCR using a 2.5х reaction mixture for PCR (cat. no. M431, Syntol, Russia). Thermocycling conditions were as follows: Preheating at 50°C for 2 min and 95°C for 10 min, followed by 40 cycles of, denaturation at 95°C for 15 sec and annealing and elongations at 60°C for 1 min. Data were analyzed using the ΔΔCq method (15). The expression of HMGA2 is specifically downregulated by hsa-let-7c (16); therefore, effective inhibition of let-7c leads to a decrease in HMGA2 mRNA level in the cells. Similarly, TWF1 expression is regulated by miR-1 and its effective transfection into the cells leads to downregulation of TWF1 (17). To perform qPCR, the isolated RNA was first converted to cDNA by RT with the Reverta kit (cat. no. K3-1-100, AmpliSens, Moscow, Russia) and then amplified with TaqMan primers for HMGA-2 (assay ID Hs00171569_m1; cat. no. 1287202; Applied Biosystems; Thermo Fisher Scientific, Inc.) or for TWF1 (assay ID Hs00702289_s1; cat. no. 4331182; Applied Biosystems; Thermo Fisher Scientific, Inc.) and endogenous control β-actin (assay ID Hs01060665_g1; cat. no. 4331182; Applied Biosystems; Thermo Fisher Scientific, Inc.).

The melanoma SK-MEL1 and BRO cell lines were transfected with relevant inhibitors or mimics and incubated in 24-well plates at 37°C in 5% CO₂ for 48 h. The cells were then detached with 0.25% trypsin solution in Hank's Balanced Salt Solution (HBSS; Gibco; Thermo Fisher Scientific, Inc.), washed twice with cold PBS and stained with the Annexin V-fluorescein isothiocyanate/7-Aminoactinomycin D (7-AAD) kit (cat. no. IM3614, Immunotech, Inc.; Beckman Coulter, Inc., Quebec, Canada) according to the manufacturer's protocol. The proportion of viable (Annexin V⁻/7-AAD⁻), early apoptotic (Annexin V⁺/7-AAD⁻), apoptotic (Annexin V⁺/7-AAD⁺) and necrotic (Annexin V⁻/7-AAD⁺) cells was detected on Cytomics FC-500 (Beckman Coulter, Inc., Brea, CA, USA) using CXP Software (version 2.2; Beckman Coulter, Inc.). The experiment was performed in triplicate.

Cell cycle analysis was performed using propidium iodide staining. Transfected SK-MEL1 and BRO cells were incubated in 24-well plates at 37°C in 5% CO₂ for 48 h. Thereafter, the cells were washed twice with PBS, then treated with RNase A (100 µg/ml) for 30 min, and stained with propidium iodide (100 µg/ml) for 30 min at 37°C in the dark. The proportion of cells in each phase was detected using a Cytomics FC-500 flow cytometer (Beckman Coulter, Inc.) using CXP Software (version 2.2).

Melanoma cells transfected with miRNA inhibitors or mimics were cultured at 37°C with 5% CO₂ for 24 h, then detached with 0.25% trypsin solution in HBSS (Gibco; Thermo Fisher Scientific, UK), and placed in 96-well plates at a final concentration of 3×10⁴ cells/ml. An MTT assay was performed at 24, 48, 72 and 96 h following transfection. For this purpose, the culture medium was replaced and 10 µl MTT solution/well was added. The cells with MTT were incubated for 4 h at 37°C with 5% CO₂. The cells were then washed, and 200 µl DMSO/well was added followed by incubation at room temperature for 10 min. Absorbance of stained supernatants was measured with Efos-9305 spectrophotometer (Shvabe Photosystems, Moscow, Russia) at a wavelength of 560 nm. Cell viability/proliferation was directly proportional to the absorbance rate. The experiment was performed in triplicate.

A total of 150 µl transfected cells at a concentration of 5×10⁴ cells/ml were placed in 96-well plates and incubated at 37°C in 5% CO₂. The cells were stained using the CyQUANT^® Direct Cell Proliferation Assay (cat. no. 35011; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol, followed by addition of NucBlue^® Live reagent from Ready Probes^® Cell Viability Imaging kit (Blue/Red) (cat. no. R37610, Molecular Probes; Thermo Fisher Scientific, Inc.). Following 30 min of incubation at room temperature, fluorescence microscopy (magnification, ×460) was performed using the Floid^® Cell Imaging Station (Floid Software, version no. 22809; Thermo Fisher Scientific, Inc.). The nuclei of proliferating cells were stained green and blue, while non-proliferating live cell nuclei were stained dark blue.

Migration and invasion experiments were performed with CytoSelect^™ 24-Well Cell Migration and Invasion assay kit (cat. no. CBA-100-C, Colorimetric Format; Cell Biolabs, Inc., San Diego, CA, USA). The 24-well migration and invasion plates contained polycarbonate membrane inserts with 8 µm pores (kit component, item no. 10001). The inserts of the invasion assay had an additional uniform layer of dried basement membrane matrix solution on the upper membrane surface (kit component, item no. 11001). Melanoma cells transfected with miRNA mimic or inhibitor were incubated at 37°C in 5% CO₂ and then detached with 0.25% trypsin solution in HBSS (Gibco; Thermo Fisher Scientific, Inc., UK), washed with PBS and suspended in FBS-free medium RPMI-1640 to a final concentration of 1×10⁵ cells/ml. Following this, the cells were placed inside each insert, while the lower well of the migration plate was filled by the medium containing 10% FBS. Following a 22-h incubation at 37°C in 5% CO₂, the cells in the interior of the inserts were mechanically removed and the rest were dissolved in the extraction solution (kit component, item no. 11003-C) of the assay for 10 min and transferred to a 96-well plate for optical density evaluation at a wavelength 560 nm using the Efos-9305 spectrophotometer (Shvabe Photosystems). Relative migration and invasion activity was determined by the ratio of the mean absorbance in the test cells vs. the mean absorbance in the negative control cells. The experiments were performed in triplicate.

The ability for colony formation was examined 24 h following transfection of the cells with miR-204-5p and miR-3065-5p inhibitors and mimics. Melanoma cells were detached with 0.25% trypsin solution and placed in a 6-well culture plate containing 2 ml culture medium RPMI-1640 with L-glutamine (Gibco; Thermo Fisher Scientific, Inc.) with 10% FBS at a concentration of 10³ cells/well. The cells were then incubated for 7–10 days at 37°C with a 5% CO₂ until they formed visible colonies containing ≥50 tumor cells. The culture medium was changed every 3 days. Following the formation of the colonies, they were washed twice with PBS, then fixed with 70% ethanol for 10 min at room temperature, then stained with 0.05% crystal violet solution (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 30 min at room temperature. The stained cells were washed four times with running water, air-dried at room temperature and the number of colonies in each well was counted by the naked eye. The experiment was performed in triplicate.

DIANA-mirPath v.3.0 was used for the KEGG pathway analysis (http://www.genome.jp/kegg; 01/01/2018) of miRNA signature. miRNA targets were predicted based on DIANA-microT-coding sequence. The threshold (P≤0.05) was calculated using the Fisher's exact test. To predict the gene potential targets of miR-3065-5p and miR-204-5-p, three different algorithms [TargetScan (version 7.0; http://www.targetscan.org), miRDB (version 5.0; http://mirdb.org/miRDB) and miRTarBase (version 4.5; http://mirtarbase.mbc.nctu.edu.tw/)] were applied to identify the validated targets of hsa-miR-3065-5p and miR-204-5-p. Subsequently, genes associated with carcinogenesis for each of the miRNAs were selected.

A Gene Ontology term enrichment analysis (http://www.geneontology.org; 14/12/2017) was performed to identify the biological processes, molecular functions and cellular components associated with the target genes of miR-3065-5p and miR-204-5p. The KEGG pathway enrichment analysis was used to identify the miRNA targets associated with signaling pathways in melanoma. P<0.05 was considered to indicate a statistically significant difference. The PANTHER^™ software classification system (version 10.0; Protein ANalysis THrough Evolutionary Relationships; www.pantherdb.org) was used to provide interpretation of the biological functions of the validated targets of miR-204-5p and miR-3065-5p.

Statistical analysis for all experiments was performed using non-parametric Mann-Whitney U-tests for two independent groups (negative control group and experimental group) with Statistica 6.1 software (StatSoft, Inc., Dell, Round Rock, TX, USA). The data are presented as the mean ± standard error of mean. miR expression levels data in melanoma, melanocytic nevi and melanoma cell lines were compared using Kruskal-Wallis H test. Multiple comparisons of mean ranks were conducted for all groups. P<0.05 was considered to indicate a statistically significant difference.

miR-204-5p and miR-3065-5p expression was evaluated in skin samples obtained from patients with melanoma and melanocytic nevi, and in the melanoma BRO and SK-MEL1 cell lines. The miR-204-5p levels were identified to be decreased in the BRO melanoma cells compared with those in the melanocytic nevi (P=0.023, respectively; Fig. 1). The expression level of miR-3065-5p did not differ between the groups analyzed.

To determine the functional role of miR-204-5p and miR-3065-5p in melanoma cells, bioinformatics analysis was performed using the aforementioned databases. A total of 235 target genes were identified for miR-204-5p. Among those, 36 were selected as being associated with carcinogenesis (Table I). Of those, the predicted targets were: Negative apoptosis regulator B-cell lymphoma 2 (Bcl-2), transcription factor Sex Determining Region Y-Box 4 (SOX4), and Forkhead box C1 (FOXC1). These genes were predicted using three databases: TargetScan, miRDB, miRTarBase. The primary signaling pathways associated with miR-204-5p were steroid biosynthesis (hsa00100), cytochrome P450 xenobiotic metabolism and apoptosis (hsa04210). The 184 genes corresponding with miR-3065-5p were detected to be associated with 48 altered biological processes, among which 13 target genes were associated with tumor growth and progression (Table II). One of those is Homeodomain-interacting protein kinase 1 gene (HIPK1), which participates in apoptosis, angiogenesis and cell proliferation (18,19). The other miR-3065-5p gene target was ITGA1 (integrin receptor subunit α1), which may regulate tumor growth and angiogenesis (20). miR-3065-5p was associated with the following signaling pathways in accordance with DIANA-mirPath v.3.0: Ubiquitin-mediated proteolysis (hsa04120), protein processing in the endoplasmic reticulum (hsa04141), lysine degradation (hsa00310), transcriptional dysregulation in cancer (hsa05202) and adherens junctions (hsa04520).

To investigate the functional role of miR-204-5p in melanoma, its specific inhibitors or mimics were transfected in BRO and SK-MEL1 melanoma cells. Transfection efficiency analysis revealed that the mimics and inhibitors of the studied miRNAs resulted in respective up- or downregulation of the target molecules (Fig. 2). miR-204-5p expression alterations caused by application of either inhibitors or mimics resulted in decreased cell viability/proliferation, as measured by the MTT assay in BRO and SK-MEL1 melanoma cells (Fig. 3). The miR-204-5p inhibitors exerted no effect on the rate of BRO melanoma cell apoptosis. Application of miR-204-5p mimics led to a decrease in the number of surviving BRO melanoma cells, but did not affect the ratio of apoptotic and necrotic cells (Fig. 4).

Application of either miR-204-5p inhibitors or mimics resulted in the increase of the percentage of cells at the G1 stage and the decrease of those at stage S-G2 among SK-MEL1 melanoma cells (Fig. 4), and miR-204-5p inhibitor significantly decreased S-G2 stage in BRO melanoma cells (Fig. 4). miR-204-5p downregulation resulted in 1.54-fold increase of cell migration but did not affect the invasion of BRO melanoma cells (Fig. 5). Transfection of the cells by mimics did not alter these characteristics (P>0.05) but reduced their colony formation ability (Fig. 6). Melanoma SK-MEL1 is a low-adhesion cell line, which becomes slightly invasive in vitro (21); therefore, its migration, invasion and colony formation were not estimated.

miR-3065-5p expression modulation by inhibitors or by mimics led to an apparent decrease of cell viability/proliferation in BRO and SK-MEL1 melanoma cells (Fig. 3). Apoptosis analysis demonstrated that all transfected cells had live and apoptotic cell ratios similar to negative controls (P>0.05). miR-3065-5p inhibition did not affect the cell cycle of either cell line, but miR-3065-5p mimics reduced the number of SK-MEL1 cells at the S-G2 phase (from 24.06±0.64 to 20.95±0.57%; P=0.0495) and increased the cell population at the G1-phase (from 74.87±0.72 to 78.21±0.54%; P=0.0495; Fig. 4). miR-3065-5p inhibition stimulated BRO melanoma cell migration, whereas miR-3065-5p upregulation exerted the opposite effect (Fig. 5). It was also identified that miR-3065-5p inhibitor or mimic application promoted invasion of BRO melanoma cells, whereas miR-204-5p mimics induced suppression of BRO melanoma cell invasive ability (Fig. 5). Upregulation of miR-3065-5p caused the change in the colony number of BRO cells (Fig. 6).

To elucidate the molecular mechanisms underlying the involvement of miR-204-5p and miR-3065-5p in melanoma cell biological behavior, the effect of these miRNAs on the expression of their target genes was investigated. Bcl-2, Transforming growth factor β receptor 1 (TGFβR1) and SOX4 gene expression levels were evaluated following performing gain- and loss-of-function experiments for miR-204-5p, and HIPK1 and ITGA1 for miR-3065-5p. The inhibition of miR-204-5p in BRO melanoma cells was identified to decrease the level of Bcl-2, while stimulation of miR-204-5p exhibited no effect on Bcl-2 expression. Conversely, Bcl-2 expression was decreased in melanoma SK-MEL1 cells following miR-204-5p mimic transfection, and remained stable following specific miR-204-5p inhibitor application. The mRNA levels of TGFβR1 were downregulated following the application of the inhibitor and mimic of miR-204-5p in BRO melanoma cells, and following miR-204-5p mimic transfection in SK-MEL1 melanoma cells. miR-204-5p inhibition did not affect TGFβR1 expression in SK-MEL1 cells. No alterations in SOX4 expression were observed following miR-204-5p inhibitor and mimic application in either cell line (Fig. 7).

Melanoma cell transfection by miR-3065-5p inhibitors and mimics led to a downregulation of HIPK1 in BRO melanoma cells and upregulation in SK-MEL1 melanoma cells. The ITGA1 level was downregulated by miR-3065-5p inhibition and upregulated by its mimics in BRO melanoma cells. On the contrary, overexpression of ITGA1 was observed following miR-3065-5p knockdown in SK-MEL1 cells, while transfection with miR-3065-5p mimics did not affect ITGA1 expression (Fig. 7).