The HEMn-LP, HEMn-MP and HEMn-DP melanocyte cell lines were purchased from Thermo Fisher Scientific, Inc., while the A375, G361, Sk-Mel-28, HT144, Hs294t, A2058, Sk-Mel-2, Sk-Mel-3, WM-266-4 and VMM1 melanoma cell lines were obtained from the American Type Culture Collection (ATCC). The Mel-Ho, Colo-800 and RPMI-7951 melanoma cell lines were acquired from Innoprot S.L.

The cell lines used in the present study may be classified into three different groups according to their origin: Normal melanocytes (HEMn-LP, HEMn-MP and HEMn-DP), primary melanomas (A375, G361, Sk-Mel-28, ME4405 and Mel-Ho) and metastatic melanomas (HT144, Hs294t, A2058, Sk-Mel-2, Sk-Mel-3, WM-266-4, VMM1, RPMI-7951 and Colo-800). Table SI provides a brief description of these cell lines.

The Mel-Ho cell line was authenticated by short-tandem repeat DNA profiling analysis (TH01: 7; D21S11: 29; D5S818: 12; D13S317: 11; D7S820: 10,11; D16S539: 11; CSF1P0: 12; AMEL: X; VWA: 14,18; TPOX: 8,10. The numbers after the loci name indicate the number of repeats of the repeat unit for that locus in the analysed cells' genome).

Melanocyte cell lines were cultured in medium 254 supplemented with human melanocyte growth supplement (Thermo Fisher Scientific, Inc.). The A375, ME4405, Hs294t, RPMI-7951, A2058 and HT144 lines were maintained in DMEM culture medium (MilliporeSigma), while the Sk-Mel-28, WM-266-4 and Sk-Mel-2 lines were cultured in Eagle's Minimum Essential Medium (ATCC). RPMI 1640-GlutaMAX™ (Thermo Fisher Scientific, Inc.) was used to culture the Mel-Ho, Colo-800 and VMM1 cell lines, and McCoy's 5A was used for G361 and Sk-Mel-3 cells (Thermo Fisher Scientific, Inc.). All the culture media used for melanoma cell lines were supplemented with 10% FBS (v/v; Cytiva), 100 UI/ml penicillin and 100 µg/ml streptomycin (Thermo Fisher Scientific, Inc.). The cells were cultured at 37°C in a humidified atmosphere with 5% CO₂. All of the cultured cell lines tested negative for mycoplasma (Venor^® GeM One Step Test Kit; Minerva Biolabs).

Cell protein extracts were obtained from cells incubated for 15 min in RIPA buffer (Thermo Fisher Scientific, Inc.) containing protease inhibitor cocktail-3 and a phosphatase inhibitor (MilliporeSigma), and then centrifuged for 5 min at 15,000 × g at 4°C. The proteins were quantified with a BCA assay (EMD Millipore) and equal amounts (40 µg) of each protein extract were denatured for 30 min at 37°C (22) and separated by 7.5% SDS-PAGE electrophoresis under reducing conditions. The proteins were then wet-transferred to a nitrocellulose membrane (Cytiva) that was probed overnight at 4°C with primary antibodies against PLA₂, PLC, PLD1, PLD2 and tubulin. Subsequently, the membranes were incubated with HRP-conjugated secondary antibodies for 2 h at room temperature. The details of the antibodies are provided in Table I. The bands were visualized by enhanced chemiluminescence and digital images of the bands were obtained with a charge-couple device camera-based imager, the Syngene G box imaging system (Syngene).

Cells were seeded on round coverslips, incubated overnight, fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X-100. The cells were incubated overnight at 4°C with anti-human PLD2 primary antibody (1:200 dilution; Abgent), washed and then incubated for 2 h at room temperature in the dark with an Alexa-Fluor 594-conjugated secondary antibody (1:5,000 dilution; cat. no. ab150080; Abcam). Finally, the cells were incubated with Hoechst-33342 (Thermo Fisher Scientific, Inc.) for 5 min and the coverslips were mounted with Fluoromount-G (MilliporeSigma) on a microscope slide. The cells were visualized by generating 20 z-stacks on a Leica SP8 confocal fluorescence microscope (Leica Microsystems GmbH) at an original magnification of ×63.

PLD activity in melanoma cells was assessed using a protocol described previously (26). First, liposomes were prepared from a short chain PC (8:0/8:0) (3.5 mM; Avanti Polar Lipids) in HEPES (pH 7.8; 45 mM), phosphatidylinositol 4,5-biphosphate (1 µM; Avanti Polar Lipids, Inc.) and 1.0 µCi [³H]-n-butanol (American Radiolabeled Chemicals Inc.). Pellets of the cell lines studied were resuspended in lysis buffer [5 mM HEPES (pH 7.8), 100 µM Na orthovanadate, 0.4% Triton X-100, 2 µg/ml aprotinin and 5 µg/ml leupeptin], sonicated on ice and the protein content of the sonicates was measured with a BCA assay. The liposomes were incubated for 20 min on a shaker with 50 µg of the cell sonicates at 30°C to achieve transphosphatidylation of PLD. This reaction was stopped by adding ice-cold chloroform/methanol (1:2) and the lipids in the samples were isolated and resolved by thin-layer chromatography. The amount of [³H]-phosphatidylbutanol (PBut) that co-migrated with the PBut standards was measured in a scintillation counter (PerkinElmer, Inc.).

Melanoma cells were seeded in 6-well plates and grown until 60–70% confluence. They were then transfected by incubation for 48 h with a transfection mixture of 1 µg of plasmid DNA (cat. no. RC202042; Origene Technologies, Inc.), TransIT^®−2020 transfection reagent (Mirus Bio, LLC) and Opti-MEM™ (Thermo Fisher Scientific, Inc.). Control cells were transfected with empty plasmid (pc-DNA3.1-myc; Origene Technologies, Inc.) following the same protocol and the overexpression efficiency was confirmed using western blot analysis.

When the melanoma cells cultured in 6-well plates reached 60–70% confluence, the culture medium was replaced with silencing mixture containing 50 nM small interfering (si)RNA (Ambion; Thermo Fisher Scientific, Inc.), Opti-MEM™ and Lipofectamine RNAiMax (Thermo Fisher Scientific, Inc.), and the cells were incubated for 48 h prior to performing the experiments. For control transfections, a non-targeting siRNA molecule suitable for negative controls (cat. no. AM4611; Ambion; Thermo Fisher Scientific, Inc.) was used following the same procedure and the correct silencing of the cells was confirmed using western blot analysis.

The relative proliferation capacity of cells after PLD2 silencing and overexpression was measured as the number of viable cells in that condition after 24 h in culture, relative to their control. Viable cells were quantified by XTT, which is based on dehydrogenase enzymes that reduce the tetrazolium salt to a highly coloured formazan dye. The amount of the formazan produced is proportional to the number of viable cells. It was confirmed that the treatments did not affect dehydrogenase enzyme activity compared to the results obtained by XTT and those calculated by trypan blue, so XTT is valid for measuring cell viability in the present study. Melanoma cells were grown overnight in 96-well plates in complete culture medium and then the medium was discarded and low-FBS medium (0.5%) was added. After 2 h in the incubator at 37°C, the medium was replaced with complete culture medium and 24 h later, XTT reagent (MilliporeSigma) was added to the cells. The cell absorbance in each experimental condition was measured at 450 nm in a Synergy HT spectrophotometer (BioTek Instruments, Inc.) and the proliferation rate of the cells was expressed as a ratio relative to their corresponding transfection controls.

Cell migration was assessed using Boyden chambers with 8-µm pores (Corning, Inc.). Melanoma cells were incubated for 2 h at 37°C in low-FBS medium (0.5%) and 400 µl of 5×10⁴ cells/ml were seeded in the upper chambers in medium containing 0.5% FBS. For invasion, 400 µl of 7.5×10⁵ cells/ml in 0.5% FBS medium were seeded in the upper chambers of Matrigel^®-coated Boyden chambers (Corning, Inc.). Complete culture medium was added to the bottom well and after 20 h of incubation at 37°C, the non-migrated/invaded cells were removed by aspiration and wiping the upper side of the membrane with a cotton bud, while the migrated/invaded cells were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, Inc.) and stained with 0.2% crystal violet at room temperature for 8 min. Finally, the filter was mounted and observed under a microscope at a magnification of ×20. Images of six different fields were acquired and the cells in each field were counted. The migratory/invasive capacity of each cell line was calculated as a ratio relative to their corresponding transfection control.

Values are expressed as the mean ± standard error of the mean of at least three experiments and the statistical significance of the differences between the means was calculated using one-way ANOVA and Bonferroni's post-hoc correction for the comparison of more than two data sets, or Dunnett's multiple-comparison test post-hoc correction for comparison of multiple data sets with their control as indicated in each experiment (GraphPad Prism 5.01; GraphPad Software, Inc.). P<0.05 was considered to indicate a statistically significant difference.

As several phospholipases are overexpressed in a variety of cancers (1,27) and they participate in their malignant transition, the presence of this family of enzymes was assessed in melanoma. The expression of PLD2, PLD1, PLA₂ and PLC was observed to vary in western blots of 17 cell lines, including skin melanocytes, as well as in primary and metastatic melanoma cell lines. In particular, while PLD2 was weakly expressed in skin melanocytes, it was expressed more strongly in melanoma cells (Figs. 1A and B and S1), particularly metastatic cells, although this increase in expression was cell line-specific. Conversely, no clear variation was observed in PLD1 protein expression between skin melanocytes and melanoma cells (Figs. 1A and C and S2). In fact, the expression of this protein in the HEMn-MP cell line was similar to that in all the primary melanomas except Mel-Ho. PLA₂ was expressed more strongly by skin melanocytes than by primary melanomas (Fig. S3), although HT144, Colo-800 and RPMI-7951 metastatic melanomas had the highest levels of PLA₂, a trait that was cell line-specific as it was not common to all the metastatic melanomas. PLC was expressed relatively homogeneously in the cell lines studied and no marked differences were observed (Fig. S3). The enhanced presence of PLD2 in melanoma cells was specific to this isoform, since the PLD1, PLA₂ and PLC proteins remained relatively constant in melanocytes and melanoma cells.

Immunofluorescence analyses confirmed that PLD2 expression was upregulated in the malignant cell lines studied (Fig. 2). PLD2 was expressed more intensely in primary melanomas (A375, Mel-Ho) and metastatic melanomas (Hs294t, Colo-800), while its expression was faint in skin melanocytes (HEMn-LP, HEMn-MP). By contrast, immunofluorescence analysis of the other phospholipases (PLA₂, PLC and PLD1) did not indicate any consistent differences between the cell lines studied (data not shown). Indeed, both the western blot and immunofluorescence data corroborated the enhanced expression of PLD2 in primary and metastatic melanomas relative to normal skin melanocytes.

A comparative analysis of PLD activity was performed in melanocytes, primary melanomas and metastatic melanomas (Fig. 3). The results indicated that melanoma cells exhibited enhanced PLD activity and that this was associated with melanoma progression. Indeed, metastatic melanoma cells had significantly more lipase activity, while primary melanoma cell lines also had higher activity than skin melanocytes. However, the activity in the HEMn-LP melanocyte cell line was comparable to that in primary melanomas and therefore, the differences were not statistically significant between both groups. These results confirmed that, overall, the lipase activity of PLD enzymes is increased in melanoma cells and raise the question of whether they have a role in melanoma, particularly in the metastatic process.

After demonstrating that aggressive melanoma cells have stronger PLD2 expression and activity than skin melanocytes, modifications to the amounts of this enzyme were performed to shed light on the implication of PLD2 in specific carcinogenic events. The effects of transient overexpression and silencing of PLD2 were examined in primary melanoma cells (A375, Mel-Ho) or metastatic melanoma cells (Colo-800), as confirmed in western blots (Figs. 4A and S4). Furthermore, these molecular alterations were also manifested through enzymatic activity, as lipase activity was altered in accordance with protein expression in all cases, although only in the A375 line, the lipase activity exhibited statistically significant differences compared to the transfection controls. As indicated in Fig. 4B, PLD2 overexpression was associated with an increase in lipase activity when the ratio relative to the transfection control was considered: A375 cells, 1.87; Mel-Ho, 1.54; and Colo-800, 1.4. By contrast, PLD2-silenced cells exhibited weaker lipase activity: A375, 0.52; Mel-Ho, 0.82; and Colo-800, 0.68.

The functional implications of the modifications to PLD2 expression were analyzed by studying the differences in proliferation, migration and invasion of transformed cells. Although these changes were cell line-specific, there was a clear trend towards an increase in proliferation, migration and invasion of PLD2-overexpressing cells, while these processes were downregulated in PLD2-silenced cells (Figs. 5 and 6). Indeed, cell proliferation increased markedly in A375 and Colo-800 cells overexpressing PLD2 but only slightly in MelHo cells. Conversely, in cells that downregulated PLD2, there was a consistent reduction in proliferation: A375, 0.55 (P<0,001); Mel-Ho, 0.8; and Colo-800, 0.79 (Fig. 5).

Melanoma cell migration and invasion were then assessed and PLD2-overexpressing primary melanoma cells exhibited moderate increments in cell migration: A375, 1.39; and Mel-Ho, 1.31. This effect was markedly enhanced in metastatic cells (Colo-800, 2.13), whereas the results were more similar among the silenced cell lines: A375, 0.66; Mel-Ho, 0.53; and Colo-800, 0.54 (Fig. 6A and B; images are provided in Fig. S5). Finally, increases in cell invasion in association with PLD2 overexpression were observed in primary melanoma cells (A375, 1.68; Mel-Ho, 1.71) relative to metastatic cells (Colo-800, 1.43). However, in the silenced cells, the results were more similar among the three cell lines, although primary melanoma cells were still more strongly affected: A375, 0.48; Mel-Ho, 0.53; and Colo-800, 0.60 (Fig. 6C and D; images are provided in Fig. S6). In conclusion, the present results suggested that higher levels of PLD2 increased the proliferation, migration and invasion of melanoma cells, while silencing of this enzyme decreased those activities.