miR-638 promotes melanoma metastasis and protects melanoma cells from apoptosis and autophagy

The present study identified miR-638 as one of the most significantly overexpressed miRNAs in metastatic lesions of melanomas compared with primary melanomas. miR-638 enhanced the tumorigenic properties of melanoma cells in vitro and lung colonization in vivo. mRNA expression profiling identified new candidate genes including TP53INP2 as miR-638 targets, the majority of which are involved in p53 signalling. Overexpression of TP53INP2 severely attenuated proliferative and invasive capacity of melanoma cells which was reversed by miR-638. Depletion of miR-638 stimulated expression of p53 and p53 downstream target genes and induced apoptosis and autophagy. miR-638 promoter analysis identified the miR-638 target transcription factor associated protein 2α (TFAP2A/AP-2α) as a direct negative regulator of miR-638, suggestive for a double-negative regulatory feedback loop. Taken together, miR-638 supports melanoma progression and suppresses p53-mediated apoptosis pathways, autophagy and expression of the transcriptional repressor TFAP2A/AP-2α.

After 24 h of transfection, the cells were either left untreated or treated with 100 nM rapamycin for another 24 h, after which they were fixed and microscopic images were taken (x60).  ANOVA was performed on the quantile normalized data of the distinctively treated cells.

Supplementary Tables Supplementary
Differentially expressed genes were considered significant when adjusted p ≤ 0.01 (corrected for multiple testing by the Benjamini and Hochberg method). The genes which were both exclusively suppressed by miR-638 overexpression and de-repressed by miR-638 knockdown were considered for further validation. Finally, we reduced the number of candidates by considering only those genes with a total log 10 fold change of 2 or more (sum of absolute fold changes). The data was processed and analyzed using the Bioconductor package lumi [2] in the statistical programming environment R.

Cell cycle analysis
Cell cycle assays were performed as previously described [3]. Briefly, SK-Mel-147 or SK-Mel-28 cells, were transfected with indicated oligonucleotides. Twenty four hours after transfection the cells were fixed with 70% ethanol. Ethanol-fixed cells were centrifuged at 3000 rpm for 5 min, washed twice with PBS and then incubated with 0.5 ml PBS containing 0.1 mg/ml RNase A and 50 μg/ml propidium iodide (PI) for 30 min at room temperature. The cell cycle distribution was analyzed using BD-FACSCalibur (BD Biosciences Pharmingen, San Diego, CA, USA).

Proliferation Assay
Twenty four hours after transfection of SK-Mel-147 or SK-Mel-28 cells with the indicated oligonucleotides or cDNA plasmids, cells were seeded in a 96-well-plate (4 x 10 3 cells/well).
Proliferation was analyzed at given time points using the Cell Proliferation Kit II (Roche Molecular Biochemicals, Mannheim, Germany). The UV absorption was measured 2 h after addition of XTT reagent at 492 nm using a Biotek Synergy TM HT microplate reader (BioTek Instruments, Inc, Vermont, USA).
Invasion assays-Assays were performed using Boyden chamber inserts coated with matrigel layer over a 8 μm porous membrane (24 well Thincert TM , Greiner Bio-One, Frickenhausen, Germany). In brief, SK-Mel-147 cells (5 x 10 3 ) suspended in serum-free culture medium were seeded in the matrigel insert. In the lower chamber culture medium contained 20% fetal calf serum. Two to 4 h after seeding, the cells were treated with 10 μM mitomycin C for 1 h to inhibit proliferation. After 48 h, the inserts were removed, washed with PBS and fixed with 4% formaldehyde. The cells remaining on the inner membrane were removed and the invaded cells stained with DAPI. Microscopic pictures were acquired at 10 X magnifications using a BZ-9000 fluorescence microscope and analyzed using a BZ-II analyzer (Keyence, Neu-Isenburg, Germany).

Colony-forming assay
SK-Mel-28-control or SK-Mel-147 cells transfected with the indicated oligonucleotides were seeded at low density (500 cells/well) in a 6-well-plate and allowed to grow for 10-12 days.
The cells were then stained with 0.1% crystal violet and analyzed microscopically. Quantification was performed by counting the total number of colonies in each well.
Immunodetection and quantification of protein bands was performed with a LI-COR Odyssey scanner and Odyssey 3.0 software (LI-COR Biosciences).

Luciferase reporter assay
Luciferase reporter gene assays were performed as described [3]. In brief, TP53INP2 3'-UTRs were cloned into the pmirGLO vector. The miR-638 binding seed regions were mutated with mutant primers (mentioned below) using a site-directed mutagenesis kit (QuickChange Animal experiments SK-Mel-147 human melanoma cells overexpressing miR-638 or scrambled-control were trypsinized and single cells were enriched by filtering them through a cell strainer. Melanoma cells (2 x 10 6 ) were then injected into the lateral tail vein of NSG (NOD scid IL2 receptor gamma chain knockout) mice [4]. On day 21 post injection, the mice were sacrificed and the lungs were dissected for macroscopic and microscopic histological analysis. After acquiring macroscopic images, the lungs were embedded into the tissue freezing medium and sectioned for subsequent immunohistochemical analysis or stored at -80 o C. Tissue sections (8 μm thickness) were stained with hematoxylin and eosin (H&E). The images were acquired using BZ-9000E microscopic system (Keyence) at 4-fold magnification, and the extent of metastasis was analysed with BZ-II analysis software (Keyence). All animal experiments were performed according to the institutional and state guidelines, and the committee of animal welfare of Saxony approved animal protocols used in this study (TVV 53/11).

Kinetic modelling of epigenetic deregulation of miR-638 promoter
The kinetic model accounts for the evolution in time of the expression levels for: miRNA-638 (represented in the model equations with the variable miR), transcriptionally active mRNA (mA) and protein (A) levels for AP-2, transcriptionally active p53 (p53 * ), MDM2 (M2) and an additional variable that accounts for expression of targets whose expression is regulated by both p53 and AP-2TgtAdditionally, the model includes two tunable parameters accounting for the level of cellular stress (CS) and the methylation level of the AP-2 binding site in the promoter region of miR-638 (MT). The model is composed of six ordinary differential equations with the following structure: In case of miRNA-638, the model contains kinetic rates accounting for its AP-2 repressed synthesis (characterized by the parameters F m, g and k 1 ) and for its basal degradation (k dm ).
For transcriptionally active AP-2 mRNA, we included kinetic rates accounting for its basal synthesis (F mA ), as well as for its basal degradation (k dmA ) and miR-638-mediated repression (k 2 ). For AP-2, the model includes kinetic terms for its basal synthesis and degradation (k dA ).
For p53, there were included terms describing the cellular stress mediated expression (CS), as well as its basal (k dp53 ) and MDM2-enhanced degradation (k m2 ). For MDM2, the model includes kinetic terms for its p53-mediated expression and its degradation (k dm2 ). For the variable Tgt, we included a kinetic term accounting for the p53 and AP-2mediatedexpression and another accounting for its degradation (k dtg ). The parameter k 1 depends on the methylation level of the AP-2 binding site (MT) according to the following function [5]: