The α-melanocyte stimulating hormone/peroxisome proliferator activated receptor-γ pathway down-regulates proliferation in melanoma cell lines

B16-F10 cell line is a classical melanoma cell line widely employed in pigmentation studies, because it expresses a wild type MC1R, with an intact transduction machinery, which is activated in response to receptor stimulation [26]. B16-F10 mouse melanoma cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with heat-inactivated 7% fetal bovine serum (FBS) and antibiotics (all products purchased by EuroClone, Milan, Italy). The human Mel13-SM melanoma cell line (Mel 13) was isolated in our laboratory as previously reported [27], exclusively from excess parts of the biopsy collected for histological examination, without compromising the standard diagnostic procedure. Mel 13 derived from a skin metastasis removed from the leg of a 52-year-old male (pT3bN0M1, stage IV). Mel 13 cell line was grown in OptiMEM (Invitrogen Life Technologies Italia, Monza, Italy) medium containing 10% FBS and antibiotics. All the experiments were performed at low cell culture passage. Two primary human melanocyte cultures, NHM 1 and NHM 2 respectively (NHMs), were set up in our laboratory from neonatal foreskin in accordance with a previously described procedure [28]. NHMs were selectively grown in the defined medium M254 (Invitrogen Life Technologies Italia, Monza, Italy) and Human melanocyte growth supplements (HMGS) (Invitrogen Life Technologies Italia, Monza, Italy). NHMs were sub-cultured once a week and experiments carried out on cells between passages 2 to 6.

B16-F10 cell cultures and human melanoma cells were plated and 24 h later were stimulated with chemicals in fresh medium. Primary human melanocytes were plated and 24 h later were stimulated with chemicals in a fresh medium deprived of bovine pituitary extract (BPE) and phorbol 12-myristate 13-acetate (PMA). The following doses of chemicals were employed: 10⁻⁷ M αMSH (Sigma-Aldrich Srl, Milan, Italy); 3 μM GW9662 (Sigma-Aldrich Srl, Milan, Italy) a potent and irreversible antagonist of PPARγ [29]; 15 μM 3 M3-FBS (3 M3) (Merck KGaA, Darmstadt, Germany), a PLC inducer [30]; 1 μM Forskolin (FSK) (Sigma- Sigma-Aldrich Srl, Milan, Italy), a cAMP-stimulating agent [8].

Institute Research Ethics Committee (IFO) approval was obtained to collect samples of human material for research. The Declaration of Helsinki Principles was followed, and patients gave written informed consent. For biopsies obtained from minors written consent was given by the parents or legal guardians.

Total RNA was isolated using the Aurum™ Total RNA Mini kit (Bio-Rad Laboratories Srl, Milan, Italy). Following DNAse I treatment, cDNA was synthesized using oligo-dT primers and using ImProm-II™ Reverse Transcriptase (Promega Corporation, Madison, WI) according to the manufacturer’s instructions. Quantitative real time RT-PCR was performed in a total volume of 15 μl with SYBR Green PCR Master Mix (Bio-Rad Laboratories Srl) and 200 nM concentration of each primer. The primer sequences were as follows: hMC1R sense: 5′-CCTGAAGACCTCACTAGG -3′ and antisense: 5′-CATCTTGTAGAGCCTGAG-3′; mMC1R sense: 5′-CAAGGAGGTGCTGCTGTG-3 and antisense: 5′-TAGACAAATGGAGATCAGGAAGG-3′, β-actin sense: 5′-GACAGGATGCAGAAGGAGATTACT -3′ and antisense 5′-TGATCCACATCTGCTGGAAGGT-3′. Reactions were carried out in triplicate using the Real-Time Detection System iQ5 (Bio-Rad Laboratories Srl) supplied with iCycler IQ5 optical system software version 2.0 (Bio-Rad Laboratories Srl). Melt curve analysis was performed to confirm the specificity of the amplified products. β-actin was used as an endogenous control.

Cells grown on coverslips previously coated with 2% gelatin on 24-well plates were treated with 3 M3 (15 μM) for 1, 3 and 6 h. For immunolabelling with anti-PPARγ rabbit antibody (1:50 in PBS) (Cell Signalling Tecnology, New England Biolabs, UK) cells were fixed in methanol for 4 min. at −20 °C. The primary antibody was visualized by using anti-rabbit IgG- Alexa Fluor 555 conjugated antibody (1:800 in PBS) (Cell Signaling, Tecnology, New England Biolabs, UK). Nuclei were visualized with DAPI (Sigma-Aldrich Srl). Fluorescence signals were analyzed by scanning cells in a series of sequential sections with an ApoTome System (Zeiss, Oberkochen, Germany); image analysis was performed by the Axiovision software (Zeiss) and 3D reconstruction of a selection of three central optical sections was shown in each image. Quantitative analysis of the extent of PPARγ/DAPI colocalization was determined using Axiovision colocalization module (Zeiss) analyzing 100 cells for each condition randomly taken from 10 different microscopic fields. Results are expressed as fold increase of colocalization signals with respect to the values obtained in untreated cells and are reported as mean values ± SD (%). Three distinct experiments were analyzed. For immunolabelling with anti-MC1R (N-19) goat antibody (1:50 in PBS) (Santa Cruz Biotechnology, INC) cells were fixed in PFA 4% for 15 min at room temperature. The primary antibody was visualized using anti-goat IgG-Alexa Fluor 488 conjugated antibody (1:800 in PBS) (ThermoFisher Scientific). Nuclei were visualized with DAPI (Sigma-Aldrich Srl).

Cells were lysed in denaturing conditions supplemented with a protease inhibitor cocktail (Roche, Mannheim, Germany). The protein concentration of extracts was estimated with Bradford reagent (Bio-Rad Laboratories Srl, Milan, Italy). Equal amounts of proteins were then separated on acrylammide SDS-PAGE, transferred onto nitrocellulose (Amersham Biosciences, Milan, Italy) and then treated overnight at 4 °C with: anti-tyrosinase antibody, anti-cyclin E antibody (1:1000; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA); anti-p27 antibody, anti-p21 antibody, anti-cyclin D1 antibody (all 1:1000; Cell Signalling Technology, New England Biolabs, UK). HRP-conjugated goat anti-mouse (1:3000; Cell Signalling Technology) or anti-rabbit IgG (1:8000; Cell Signalling Technology) were used as secondary antibodies. Antibody complexes were visualized using the an enhanced chemiluminescence reagent (ECL) (Amersham Biosciences, Milan, Italy). A subsequent hybridization with anti-GAPDH or anti-HSP70 (both 1:5000; Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) was used to estimate the equal protein loading. Densitometric analysis was performed using a GS-800 Calibrated Image Densitometer (Bio-Rad Laboratories Srl, Milan, Italy) or with UVITEC Imaging System (Cambridge, UK). Results refer to three independent experiments.

Cells were plated in a 12 well plate at a density of 1 X 10⁴ cells/well and were grown overnight. Cells were treated with 10⁻⁷ M αMSH, 3 μM GW9662, 15 μM 3 M3 and 1 μM FSK at different time points, in accordance with the experimental design. Cell number was determined by cell counting using a phase contrast microscope. None of the employed doses of chemicals determined positivity to the Trypan blue exclusion assay test. The results presented are the average of three experiments in triplicate.

Cells were harvested by treatment with 0.25% trypsin, fixed with ice-cold 70% ethanol solution, and stained in PBS containing propidium iodide (100 μg/ml) and RNase A (90 μg/ml) overnight at 4 °C. The DNA content of the cells was measured by FACScalibur flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). A total number of 10⁴ cells for each sample were acquired. Flow cytometric histograms were analyzed by defining borders between pre-G₁, G₁, S and G₂ + M phase populations. Cell Quest Software was used to analyze data.

Cells were transfected with pGL3-(Jwt) 3TKLuc reporter construct [31] using specific Amaxa® Nucleofector kits (Lonza, Basel, Switzerland) according to the manufacturer’s instructions. 24 h after treatment with 10⁻⁷ M αMSH or 15 μM 3 M3, cells were harvested in 100 μl lysis buffer and 20 μl of the extract were assayed for luciferase activity using Promega’s Dual Luciferase (Promega Corporation, Madison, WI, USA) according to the manufacturer’s protocol. The renilla luciferase plasmid was also transfected as an internal control for monitoring transfection efficiency and for normalizing the firefly luciferase activity. The luciferase activity was expressed as fold of the activity obtained in cells treated with the different molecules divided by luciferase activity from non-stimulated cells. The pGL3-(Jwt)3TKLuc reporter construct is described elsewhere [31]. The results represent the average of three independent experiments performed in triplicate.

We investigated the capacity of αMSH to modulate proliferation in primary human melanocytes NHM 1 and NHM 2, in B16-F10 murine melanoma cell line and in the Mel 13 human melanoma cell line. All these cell lines were characterized by the genetic nature of MC1R, expressing a wild type MC1R and responding to αMSH in terms of pigmentation (see Additional file 1 and Additional file 2: Figure S1). The duplication time of primary cultures of human melanocytes is significantly longer with respect to that of B16-F10 and Mel 13 melanoma cell line (30/40 h vs 18/24 h). Thus, we planned to perform evaluations at suitable times to highlight variations in treated samples in comparison with untreated ones. The exposure to 10⁻⁷ M αMSH for 3 and 6 days, significantly increased the cell number in both NHMs (Fig. 1a). Accordingly, 48 h hrs of treatment with the hormone significantly decreased the proportion of cells distributed in the G0/G1 phase of the cell cycle (Fig. 1b). Instead, αMSH was unable to act as a mitogenic agent either in B16-F10 or Mel 13 cell lines, and it even reduced the cell number. This effect was already significant after 24 h of treatment (Fig. 1c). In compliance with these results, 24 h of treatment with αMSH determined a significant increase in the proportion of cells which were distributed in the G0/G1 phase of the cell cycle (Fig. 1d). In order to corroborate the αMSH effect on proliferation slow down, we evaluated the expression levels of G1 phase related proteins (p27, p21, cyclin D1 and E), in response to the hormone exposure. In preliminary experiments, expression levels of p27, p21, cyclin D1 and E, after exposure to αMSH, or alternative stimuli (see below), were followed at 4, 6, 12, 24, 36 and 48 h (data not shown). Here, and in the subsequent experiments (also see following paragraphs), we reported data associated with times of maximal proteins modulation. In both melanoma cell lines, the expression level of the cyclin-dependent kinase inhibitor p27 already resulted significantly up-regulated after 4 h of treatment with αMSH. The same treatment determined a significant induction of the expression level of the cyclin-dependent kinase inhibitor p21 after 48 h. Consistently, αMSH exposure significantly down-regulated the expression of both cyclin D1 and E after 6 h (Fig. 1e). Thus, these results indicate that αMSH exerted a mitogenic effect in primary human melanocytes while it reduced melanoma cell proliferation.

The cAMP/PKA pathway is the pathway through which the MC1R is universally known to promote melanin synthesis [2, 5]. FSK, as a cAMP elevating agent, mimics the effects of αMSH on melanogenesis [8]. Furthermore, FSK is employed in defined cell culture media to promote the proliferation of melanocytes [18‐20]. Here, we verified whether FSK was able to mimic the effects induced by αMSH on proliferation in NHMs. In both NHMs, the treatment for 3 and 6 days with 1 μM FSK promoted a significant increase in cell number, similar to that produced by αMSH in the same experimental conditions (Fig. 2a). In compliance with these results, the exposure to FSK promoted a significant decrease in the proportion of cells distributed in the G0/G1 phase of the cell cycle (Fig. 2b). Thus, these results indicate that the αMSH-dependent mechanism was able to promote hyper-proliferation in NHMs can be recognized in the cAMP/PKA pathway.

In our previous study on B16-F10, in a preliminary experiment of cell counting, we demonstrated that αMSH induce a slow down of proliferation. This effect was mediated by the αMSH/PPARγ pathway, a pathway independent of the cAMP/PKA canonical pathway [7]. Here, we proposed to exclude the involvement of the cAMP/PKA pathway in mediating the proliferation slow down induced by αMSH in B16-F10 and Mel 13 melanoma cell lines. Both in B16-F10 and Mel 13, the treatment for 24 an 48 h with 1 μM FSK caused an increase in cell number, producing an opposite effect in comparison to that promoted by αMSH (Fig. 2c). The analysis of cell cycle phase distribution, after 24 h of treatment with FSK, corroborated these results (Fig. 2d). Moreover, a Western blot analysis of cyclin D1 showed a rapid induction of this key regulator of cell cycle progression (Fig. 2e). These results, confirmed our initial hypothesis. In fact, they excluded the involvement of the cAMP/PKA pathway in mediating the αMSH-mediated slow down of proliferation in B16-F10 and Mel 13 melanoma cell lines.

In an attempt to find an alternative pathway which could justify the proliferation slow down induced by αMSH in B16-F10 and Mel 13 melanoma cell lines, our attention focused on the PI(4,5)P2/PLC pathway [7]. PPARγ is a tumour suppressor gene [9, 10]. Here we show that the treatment with αMSH promoted a significant induction of PPARγ transcriptional activity after 24 h of exposure, in B16-F10, Mel 13 and NHMs. Moreover, this effect was consistently higher in melanoma cell lines compared to NHMs (Fig. 3). The αMSH/PPARγ connection is mediated by the PI(4,5)P2/PLC pathway. In turn, this pathway can be directly activated by 3 M3 [30]. Thus, we tried to understand whether the treatment with 3 M3 was able to mimic the down-proliferation induced by αMSH in B16-F10 and Mel 13 melanoma cells.

Preliminary checks: first, we verified the capacity of 3 M3 to promote: (a) calcium fluxes (Additional file 3: Figure S2), (b) PPARγ translocation into the nucleus (Fig. 4a–c) and (c) PPARγ transcriptional activity by luciferase assay (Fig. 4d). (a) Concerning the analysis of calcium fluxes, the cells previously loaded with fluo-3-AM, responded to 3 M3 treatment with an increased intra-cytoplasmic calcium, which was significant at all time points evaluated. This increase consisted in an initial response, which reached a peak after 2 min, followed by a subsequent decline to plateau value after 10 min (Additional file 3: Figure S2). (b) We also evaluated the PPARγ nuclear translocation in response to 3 M3 at different time points: 1, 3 and 6 h, by immunofluorescence analysis. We observed a positive signal for PPARγ in both cytoplasm and nucleus, in untreated cells. We observed an increase of PPARγ nuclear staining in cells exposed to 3 M3, demonstrating a translocation of this transcription factor from the cytoplasm to the nucleus (Fig. 4a,b). This observation was also confirmed by the analysis of the extent of PPARγ/DAPI colocalization signals in the nucleus (Fig. 4c). (c) Luciferase assay, using the pGL3-(Jwt)3TKLuc reporter construct [31], consistently showed a significant induction of the PPARγ transcriptional activity after 24 h of 3 M3 treatment (Fig. 4d).

Then, we tried to understand whether the treatment with 3 M3 was able to reproduce the effects of αMSH on cell number and proliferation in B16-F10 and Mel 13 cells. The exposure to 3 M3 for 24 and 48 h determined a significant decrease of cell number in both melanoma cell lines. This decrease was fully comparable to that promoted by αMSH (Fig. 5a). Furthermore, the exposure to 3 M3 for 24 h, promoted a significant increase in the proportion of cells distributed in the G0/G1 phase of cell cycle (Fig. 5b). The expression level of p27 was already significantly induced after 4 h of treatment with 3 M3, in both melanoma cell lines. The same treatment determined a significant up-regulation of p21 protein expression after 48 h. Consistently, 3 M3 exposure significantly down-regulated the levels of both cyclin D1 and cyclin E after 6 h (Fig. 5c). Thus, the direct induction of the pathway by 3 M3 is able to reproduce the αMSH mediated effect on proliferation in B16-F10 and Mel 13 melanoma cell lines.

In order to strengthen the role of PPARγ in mediating the reduction in the number of cells induced by the stimulation of “PI(4,5)P2/PLC pathway”, both B16-F10 cells and Mel 13 human melanoma cells were stimulated with 3 M3, for 24 and 48 h, in the presence or absence of GW9662, as an inhibitor of PPARγ [29]. This pattern of stimulation was compared to that represented by cells treated with αMSH, in the presence or absence of GW9662. In both cell lines, the exposure of cells to 3 M3 or αMSH led to a significant decrease of number of cells. Whereas, the same treatments, in the presence of GW9662, significantly hindered the effectiveness of 3 M3 or αMSH in reducing the cell number after 24 h of treatment (Fig. 6a). Consistently with these results, the exposure of B16-F10 and Mel 13 for 24 h to 3 M3 or αMSH, was able to promote a significant increase in the proportion of cells distributed in the G0/G1 phase of the cell cycle and the co-treatment with GW9962 counteracted the effects of both the agents (Fig. 6b). Accordingly, the inhibition of PPARγ by GW9662 reverted the effects on cell cycle modulators (p27, p21, Cyclin D1 and cyclin E) promoted by αMSH in both melanoma cell lines (Fig. 6c). All together, these results underline the key role of PPARγ in mediating the αMSH dependent effect on proliferation slow down.