miR-125b induces cellular senescence in malignant melanoma

Human MM cell line Mel-Juso (DSMZ, Braunschweig, Germany) was grown in DMEM Glutamax (Invitrogen, Carlsbad, CA) with 10% FBS (Life Technologies, LT) in 37°C and 5% CO₂. Transfected Mel-Juso cells were cultured in selection medium, which was the same medium supplemented with 10-μg/mL blasticidin (Invitrogen).

Mel-Juso cells were transfected with a miRVector plasmid containing a miR-125b-1 [UCSC Genome Bioinformatics: uc010rzr.1] [37] insert and a blasticidin resistance gene (miRVec-125b) [38] or the control miRNA Vector plasmid with a blasticidin resistance gene but without any insert (miRVec-control) (Source BioScience, LifeSciences, Nottingham, UK) (map and sequence of miR-125b, see Additional file 1: Figure S1). 300.000 cells were seeded out 24 h before transfection to reach a confluency of 70–90%. 1 μg of the plasmid was mixed with 400 μl OPTI-MEM (Invitrogen) and 5 μl Lipofectamine RNAiMax (Invitrogen). The mixture was incubated at room temperature (RT) for 20 min and added to freshly washed Mel-Juso cells in 1600 μl culture medium for 24 h followed by a 14-day-culture in the selection medium in 37°C/5% CO₂. miR-125b expression was confirmed using PCR in each clone. Optimisation in the beginning of the study showed that the transfection efficacy in Mel-Juso cells was higher with Lipofectamine RNAiMax than any other transfection reagents. Transfection procedure was done in four replicates for each plasmid and each evaluated using PCR.

For the measurement of BrdU incorporation, Mel-Juso cells transfected with miRVec-125b or the miRVec-control were pulsed for 20 min with 10 μM BrdU, fixed in 70% ethanol for at least 2 h, washed with phosphate-buffered saline (PBS), incubated in 2 N HCl for 30 min, washed in neutralizing buffer (0.2 M Na₂B₄O₇, pH 8.5) and resuspended in dilution buffer (0.5% Tween 20 and 0.5% bovine serum albumin in PBS) prior to the addition of anti-BrdU mouse antibodies (Becton Dickinson, Franklin Lakes, NJ). Goat anti-mouse antibodies labeled with Alexa-Fluor 488 (1:1000; Invitrogen, Carlsbad, CA) were applied as secondary antibodies. DNA was stained with 0.0003% (30–40 μl per samlpe) 7-amino-actinomycin D solution (7AAD; Beckman Coulter, Brea, CA). Samples were analyzed by means of flow cytometry (Beckman-Coulter, Fullerton, CA). BrdU-positive events (FL1) were plotted versus cellular DNA content (FL3).

Mice experiments were performed in accordance with the national law for the experiments on animals of 14 Jun 2011 (full text available on https://​www.​retsinformation.​dk/​Forms/​r0710.​aspx?​id=​145380) and supervised by an authorised researcher (CML). All mice work is approved by Dyreforsoegstilsynet (generel permit for our laboratory work in mice, permit number 2012-15-2934-00419). All surgery was performed under sodium pentobarbital anesthesia, and every effort was made to minimize suffering.

Six Male BALB/C nude mice (Taconic, Rye, Denmark), 7 weeks of age, were housed in separate boxes with free access to water and standard laboratory food, in an animal facility at a 12 h light/dark cycle and the temperature 23°- 24°C. Mice were acclimatized for one week before the experiments. Subsequently they were anesthetized with 0.05 ml HypDorm given subcutaneously and 5 x 10⁶ Mel-Juso cells transfected with miRVec-125b vector or the control vector in 0.2 ml PBS were injected subcutaneously into the left or right flank, respectively. Tumours with a diameter of at least 1 mm were mapped separately for each animal and followed until these reached a diameter of 12 mm or for a maximum time of 4 weeks. Animals were killed and tumours were weighed and cut into three parts, which were fixed in RNA Later (for RNA isolation), 4% formalin (for standard histology, immunohistochemistry and in situ hybridization) or were frozen at −80°C (for SA-beta-gal staining). Formalin fixed samples were embedded in paraffin and serial sections were stained with hematoxylin and eosin, Ki-67 and cyclin D1, according to standard histopathology protocols. The quantification of Ki-67 and cyclin D1 expression was done semiquantitatively in each tumour using the following ordinal scale: grade 1 (0%-25% positive tumour cells), grade 2 (26%-50% positive cells), grade 3 (51%-75% positive cells), grade 4 (76%-100% positive cells). Lungs and liver were taken out and cut in to three parts each, which were treated as the tumour samples and checked for distance metastases.

Formalin-fixed and paraffin-embedded human tissue samples from the archives of the Department of Pathology (Rigshospitalet, Copenhagen, Denmark) were stage T1-T4 cutaneous MM and matched lymph node metastases (14 males and 3 females; mean age 52 years, range 24 to 83 years); mean Breslow depth 1.53 mm (range 0.37 to 4.00 mm); 15 superficial spreading MM and 2 nodular MM. Mouse tumour samples were prepared as described above. Paraffin sections (6 μm) were mounted on SuperFrost®Plus slides (Dako, Glostrup, Denmark) and air-dried for 1–2 h at RT. Sections were deparaffinized in xylene and hydrated through decreasing ethanol concentrations into PBS. Proteinase-K treatment 15 μg/ml in PK-buffer (5 mM Tris–HCl, pH 7.4, 1 mM EDTA, 1 mM NaCl) was performed at 37°C for 8 min in a volume of 300 μL in a Dako hybridizer (Dako). Sections were washed twice in sterile PBS and immediately dehydrated through an increasing gradient of ethanol solutions. Aliquots of 3 different Mercury LNA miRNA Detection Probes; hsa-miR-125b, scrambled probe, and U6 snRNA (Exiqon, Vedbaek, Denmark) were then denatured by heating to 90°C for 4 min and diluted to 40 nM, 40 nM, and 1 nM, respectively, in a formamide-free ISH buffer (Exiqon). 50 μL probe mixture was hybridized with the tissue sections in the hybridizer at 55°C for 60 min. The slides were placed at RT in 5× saline-sodium citrate (SSC) (Invitrogen) and stringency washes were performed for 5 min each at 55°C in 5× SSC (one wash), 1× SSC (two washes) and 0.2× SSC (two washes). The sections were then washed in PBS and blocked with DIG Wash and Block Buffer Set (Roche, Mannheim, Germany). Alkaline phosphatase (AP)-conjugated anti-DIG (Roche) was diluted 1:800 in the blocking solution and incubated for 60 min at RT. Slides were washed twice with PBS containing 0.1% Tween-20. Ready to use tablets (Roche) of 4-nitro-blue tetrazolium chloride (NBT) and 5-brom-4-chloro-3′-indolyl-phosphate (BCIP) substrate were dissolved in aqueous 0.2 mM levamisole. Slides were incubated for 120 min at 30°C to develop the dark-blue NBT-formazan precipitate. Sections were washed twice for 5 min in KTBT buffer (50 mM Tris–HCl, 150 mM NaCl, 10 mM KCl), and then twice in water, dehydrated in the ethanol gradient and mounted.

Monolayers of Mel-Juso cells transfected with miRVec-125b or the miRVec-control after a 24-h culture in 4-chamber glass (Nunc, Rochester, NY) or cryostat tissue sections from mice mounted on SuperFrost®Plus slides were washed twice in PBS and fixed in 0.5% glutaraldehyde (pH = 7.0) for 15 min at RT. The samples were washed twice with 20 ml 1 mM MgCl₂in PBS (pH = 6.0) and stained with the SA-beta-gal staining solution containing 1 mg/ml of 5-bromo-4-chloro-3-indolyl beta-galactopyranocid, 4% dimethylformamide, 0.0012 mM potassium ferrocyanide and 1 mM MgCl₂ in PBS (pH = 6.0) for 4 h (Mel-Juso) or overnight (melanocytes) at 37°C [3, 5]. Percentage of positive cells was evaluated blindely by one of the investigators (RG).

Cells were washed in PBS and lysed in the sample buffer 0.5 M Tris–HCl pH 6.8; 5% glycerol; 10% SDS; DTT 0.2 M) supplemented with protease inhibitor cocktail (Roche). Equal amounts of protein were separated by a 4-8% and 12% Bis-Tris gel electrophoresis at 200 V followed by electrophoretic transfer to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA). Membranes were blocked for 1 h at 4°C with Li-Cor blocking agent (Li-Cor, Lincoln, NE) before incubation with the primary antibodies against β-actin (mouse) (Sigma Aldrich, St. Louis, MO), p16 (mouse) (BD Pharmingen, San Diego, CA), p21 (mouse) (Dako), p27 (mouse) (Santa Cruz Biotechnology, Santa Cruz, CA) or p53 (rabbit) (Dako) overnight at 4°C. Subsequently, they were incubated for 1 h with the appropriate secondary antibodies labeled with 800IR dye (anti-rabbit) (Li-Cor), Alexa Fluor 680 (anti-mouse or anti-rat) both from Molecular Probes (Invitrogen). Protein bands were detected and quantified with the infrared Odyssey imaging System (Li-Cor). Quantified intensities were adjusted to the relevant actin intensity and control and sample was then compaired.

10.000 Mel-Juso cells transfected with miRVec-125b or the miRVec-control were seeded in 10 cm petri dishes containing 10 ml selection medium and cultured for 10–14 days until the appearance of macroscopically visible colonies. The plates were washed, fixed in paraformaldehyde for 24 h and stained with crystal violet 0.05% (Sigma Aldrich) for 20 min before washing in water. Plates were photographed (microscope Olympus IX 70 and camera Nikon D60) and colonies were counted manually.

1.500.000 Mel-Juso cells transfected with miRVec-125b or the miRVec-control were seeded in 10 cm Petri dishes containing 10 ml selection medium, the medium changed after 24 h and all (floating and adherent) cells collected after another 24 h. Unfixed cells were stained simultaneously with FITC-Annexin-V and PI, according to the manufacturer’s protocol (Beckman-Coulter, Fullerton, CA) and analyzed by means of flow cytometry (Beckman-Coulter, Fullerton, CA) as described previously [39].

Mel-Juso cells transfected with miRVec-125b or the miRVec-control were collected and washed twice in PBS. Murine tumour tissue samples were homogenized using a TissueLyzer II (Qiagen, Valencia, CA, USA) and processed as previously described [40]. Small RNAs were purified using the mirVana Isolation Kit (Ambion, Foster City, CA) and the PureLink RNA Micro to Midi Kit (Invitrogen) according to the manufacturer’s instructions. The concentration of RNA was measured spectrophotometrically using NanoDrop ND-1000 (Thermo Scientific, Wilmington, DE), and RNA integrity was confirmed with Agilent 2100 Bioanalyzer using Agilent Nano RNA kit (Agilent Technologies, Santa Clara, CA). miR-125b expression was measured in the samples containing the same amount of RNA with quantitative qRT-PCR assay (TaqMan TM microRNA Reverse Transcription kit-4366596, and TaqMan Universal PCR Master mix-4324018, Applied Biosystems, Foster City, CA) and validated primer sets (Applied Biosystems) according to the manufacturer’s instructions. MiR-191 was used as a reference for the normalization of qRT-PCR-data. The q-PCR was performed in triplicates using a 7900HT Fast Real-Time PCR System (Applied Biosystems).

DNA was extracted from transfected Mel-Juso cell lines containing either miRVec-125b or miRVec-control using QIAamp DNA mini-Kit (Qiagen) in accordance with the manufacturer’s protocol. PCR was made with reagents from Life Technologies as follows: Master mix (10% Gene amp PCR buffer II, 2.5 mM MgCl₂, 400 μM dNTP, 0.2 μM of each primer and 1U Ampli Taq gold) was prepared and 7.2 μL was mixed with 16.8 μL of ddH₂O and 1 μL of 5 ng/μL DNA (template). PCR was setup on an Eppendorf MasterCycler gradient cycler (Hamburg, Germany) with the following program: 95°C 10 min, then 35 cycles of: 95°C 30 s, 62°C 30 s and 72°C 1.5 min; ended with 72°C 10 min and 4°C hold. The PCR product was visually inspected on a 2.2% agarose gel on the Flashgel system (Lonza, Basel, Switzerland). The remaining PCR product was purified on Qiaquick columns (Qiagen) according to manufacturer’s specifications and eluted in 50 μL buffer. Four μL of purified PCR product were used as template in the sequencing reaction.

Sequencing was performed in duplicates with the BigDye V.1.1 cycling sequencing kit (Life Technologies) Samples consisted of 4 μL of Ready reaction mix, 2 μL of BigDye buffer, 3.2 pmol of primer (forward-GCGTTTAAACTTAAGCTTGGTACCGAGC, reverse-CATTCCCCCCTTTTTCTGGAGAC), 4 μL of template and 20 μL of ddH₂O. Reactions were performed with forward and reverse primers in duplicate with the following program: 96°C 1 min, then 25 cycles of: 96°C 10s, 50°C 5 s and 60°C 4 min, ending with 4°C hold. PCR products were purified with Centrisep columns (Princeton Separations, Adelphia, NJ) according to the manufacturer’s recommendations. Two μL of purified sequencing product were mixed with 8 μL of HiDi formamide and run on an ABI3130xl DNA Sequencer (Life Technologies) with the following specifications: Array 36 cm, polymer POP4, injection voltage 3 kV, injection time 7 s, run voltage 15 kV and run time 1700 sec. One of the reverse duplicates was removed due to low quality. Results were analyzed and assembled with Sequencher v.5 (Gene Codes Corporation, Ann Arbor, MI, USA). Differences in the chromatograms were visually inspected. Low quality regions resulting in differences were removed. Sequence ends had to be cropped due to low reaction quality.

We have previously shown by miRNA array approach that miR-125b is expressed in primary cutaneous MM [26, 41]. By comparing metastasizing and non-metastasizing stage T2 cutaneous MM we detected an overall decrease in miR-125b expression in metastasizing tumours [26]. Here we employed ISH to further define the expression pattern of miR-125b in MM. As shown in Figure 1 miR-125b is expressed in MM cells both in the primary tumours and in the sentinel node metastases. The staining of miR-125b seems to be located predominantly to the nuclei. We noted that the expression was not homogenous and some cells characterized by large size and abundant cytoplasm expressed higher amounts of miR-125b than other cells (arrows in Figure 1A,B). These larger cells could represent a more malignant, highly atypical tumour cell population, but could also comprise the population of cells undergoing spontaneous cellular senescence. Since SA-beta-gal staining is not possible on paraffin-embedded material and fresh samples from primary cutaneous MM are not available due to ethical considerations, we decided to examine the potential functional involvement of miR-125b in MM cell proliferation and senescence in vitro.

Mel-Juso cell line is established from the primary tumour of a 58-year-old woman with MM in 1977. This line has wildtype BRAF, is poorly differentiated [42] and exhibits intermediate invasiveness [43]. Transfection with miRVec-125b resulted in an 8.0 ± 1.13-fold upregulation of miR-125b by RT-qPCR compared to the miRVec-control transfected cells. After the experiments were completed, the miR-125b insert was sequenced to confirm that no mutation had occurred in the miR-125b sequence of the insert throughout the experimental period. For full insert consensus sequence, see Additional file 1: Figure S2.

As shown in Figure 2A,B and E the miRVec-125b transfected cells formed smaller and fewer colonies than the control cells (930 ± 105.2, n = 4 replicates vs. 1207.5 ± 218.8, n = 4 replicates, p-value 0.011). The miRVec-125b transfected cells showed a G0/G1 cell cycle arrest, demonstrated in Figure 2 (F, G) by a significantly increased proportion of G0/G1 cells (p < 0.05) and a significant decrease in BrdU incorporation (p < 0.05).

Microscopic examination of the colonies (Figure 2C,D) revealed that the miRVec-125b-transfected cells were enlarged, flattened, and had abundant cytoplasm, consistent with cellular senescence, in contrast with the normal spindle-shaped cell morphology in the control group. SA-beta-gal staining showed increased expression of SA-beta-gal in estimated 60-70% of the miRVec-125b transfected compared to estimated 5-10% of the control cells (Figure 3A,B). Western blot analysis of the miRVec-125b cells showed an up regulation of the expected markers of senescence: p53, p21 and p27 compaired to the control. p16 was not expressed in the Mel-Juso cell line. (Figure 3C).

To evaluate whether apoptosis was responsible for the apparent lesser growth observed in the colony forming assay we compared the proportions of annexin-positive cells in the miRVec-125b transfected cells with the controls. We found no evidence of apoptosis in either group (95.5% ±0.7% annexin-negative, viable cells in the miRVec-125b transfected cell line versus 94.7% ±1.2% in the control; p = 0.46).

To further investigate the significance of miR-125b in the regulation of MM growth we established an in vivo model of tumour formation from Mel-Juso cells injected subcutaneously into immunodeficient, nude BALB/c mice. In this model the Mel-Juso cells formed subcutaneous tumours within a time span of 4 weeks, but did not produce distant metastases (confirmed by macroscopic and histological examination of lung and liver tissue). ISH revealed considerable expression of miR-125b in the tumours originating from the miRVec-125b vector-transfected cells, in contrast to a negligible expression in the tumours emerging from the miRVec-control cells. The tumours were very homogeny in their appearances which was consistent with the fact that they were developed from a clone rather than the normal progression from normal tissue to tumour with the following heterogeneity otherwise known from Melanoma tumours [44, 45]. The stability of miR-125b overexpression was confirmed by RT-q-PCR quantification of miR-125b in the tumour tissue showing an 11.6 ± 6.0 fold increased expression in the miRVec-125b transfected tumours. As could be expected from the in vitro experiment, SA-beta-gal staining revealed a focal increase in the staining suggesting accelerated senescence in the tumours originating from miRVec-125b -transfected cells. This was paralleled by the marked decrease in the expression of proliferation markers Ki67 and cyclin D1 (Figure 4). However, we did not detect any difference in weight between the tumours originating from miRVec-125b -transfected cells and the control, miRVec-control transfected cells (mean weight miR-125b: 0.021 g ± 0.003, control: 0.023 g ± 0.004).