Calcium acts as a central player in melatonin antitumor activity in sarcoma cells

MG63 (osteosarcoma) and sw-1353 (chondrosarcoma) cell lines were purchased from the American Type Culture Collection (Teddington, UK) and cultured in DMEM (Sigma- Aldrich, St Louis, MO, USA) medium supplemented with 10% v/v Fetal Bovine Serum (FBS; Gibco, Invitrogen Life Technologies, Barcelona, Spain), 100 U/ml penicillin and 100 μg/ml streptomycin (Gibco, Spain) in a humidified atmosphere of 5% CO₂ at 37 °C. Cells were sub-cultured once a week using a 0.25% trypsin solution. Compound concentrations used for the experiments were: 1 mM melatonin, 5 µM BAPTA, 3 mM calcium chloride, 100 µM trolox and 100 µM ascorbate (Sigma-Aldrich, St Louis, MO, USA). Culture flasks and dishes were obtained from Fisher Scientific (Madrid, Spain). Fluo-3AM and Rhod-2 were purchased from Invitrogen (Invitrogen Life Technologies, Barcelona, Spain) and BAPTA/AM was acquired from Santa Cruz (Santa Cruz Biotechnology, Dallas, TX, USA). All other reagents were purchased from Sigma (Sigma-Aldrich, St Louis, MO, USA), unless otherwise indicated.

The effect of melatonin or other compounds on cell viability was evaluated using a colorimetric assay, 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). Briefly, cells were seeded in 96-well plates and once the treatments were completed, 10 µl of MTT solution in PBS (5 mg/ml) was added. After 4 h of incubation at 37 ºC, an equal volume of lysis solution [sodium dodecyl sulphate (SDS) 20% and dimethylformamide pH 4.7, 50%] was added. Next, the mixture was incubated at 37 ºC overnight and the samples were measured in an automatic microplate reader (µQuant, Bio-Tek Instruments, Inc., Winooski, VT, USA) at a wavelength of 540 nm.

To measure cytosolic calcium levels (Fluo3-AM), intracellular oxidants (DCFH-DA) and mitochondrial membrane potential (Rhodamine 123), cells were seeded in 6-well plates and after treatment incubated with the proper fluorescence probe during 20 min at 37 °C in the dark: 1 mg/ml Fluo3-AM, 10 μM DCFH-DA in serum-free medium or 1 μg/ml rhodamine 123 in serum-free medium. Next, the cells were resuspended in 500 μl PBS and incubated with propidium iodide (final concentration 2 μg/ml) for 10 min at room temperature in the dark. Fluorescence of 10,000 live cells per group (cells negative for propidium iodide staining) was measured in a Beckman Coulter FC500 flow cytometer (Beckton Dickinson, Franklin Lakes, NJ, USA).

To analyze cell cycle progression, cells were seeded in 6-well plates and after treatment harvested and fixed in TPI overnight at 4 °C [0.0025 g propidium iodide, 50 µl Triton X-100 Detergent, 0.05 g sodium citrate in 50 ml PBS]. Next, fluorescence was measured in a Beckman Coulter FC500 flow cytometer (Beckton Dickinson Franklin Lakes, NJ, USA).

Cells were seeded in 6-well plates and after treatment collected by trypsinization. Next, they were incubated with a fluorescent probe specific for mitochondrial Ca²⁺ [3 µM Rhod-2] for 30 min at 37 °C. The fluorescence signal from the cells was measured using a microplate fluorimeter FLX-800 (Bio-Tek Instruments, Inc., Winooski, VT, USA) at excitation and emission wavelengths of 552 nm and 581 nm, respectively. The obtained fluorescence was normalized to the total number of cells.

Cells were seeded in 24-well plates after which total LDH activity was determined following the specifications of the lactic dehydrogenase-based In Vitro Toxicology Assay Kit (Sigma-Aldrich, St Louis, MO, USA). Absorbance was determined using an automatic microplate reader (μQuant; Bio-Tek Instruments, Inc., Winooski, VT, USA) at 490 nm after which the obtained data were normalized to total protein concentrations.

Cells were seeded in 150 mm plates after which lactate concentrations were determined using a Lactate Assay Kit according to the manufacturer's protocol (Sigma-Aldrich, St Louis, MO, USA). Absorbance was measured at 570 nm using an automatic microplate reader (μQuant; Bio-Tek Instruments, Inc., Winooski, VT, USA).

To evaluate cell migration, ‘wounds’ were made in confluent 6-well plates by scratching across each monolayer using a 100 μl pipette tip. Next, the cultures were treated with melatonin, calcium chloride (3 mM final concentration) or a combination of both. Every hour an optical microscopic image of the scratched area was taken starting at time 0 h and extending to 24 h using a 10 × objective.

For protein expression analysis, cells were seeded in 100 mm plates. After treatment, the cells were lysed with ice-cold lysis buffer (150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% v/v Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na3VO4, 1 μg/mL leupeptin, 2 μg/ml aprotinin, 1 μg/ml pepstatin-A, 110 nM NaF, 1 mM PMSF, 20 mM Tris–HCl pH 7.5). Next, 30 μg total protein samples were separated by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membranes (Amersham Bioscience, Pittsburgh, PA, USA). The resulting blots were incubated overnight at 4 ºC with appropriate antibodies: anti-Cyclin D1 (1:1000, Cell Signaling, Danvers, MA, USA), anti-p-AKT (1:1000, Cell Signaling, Danvers, MA, USA), anti-p-ERK (1:1000, Cell Signaling, Danvers, MA, USA), anti-N-Cadherin (1:250, Santa Cruz Biotechnology, Dallas, TX, USA), anti-Vimentin (1:1000, Santa Cruz Biotechnology, Dallas, TX, USA), anti-PDH (1:500, abcam, Cambridge, UK) and anti-β-tubulin or anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as loading controls (1:5000, Santa Cruz Biotechnology, Dallas, TX, USA). Immunoreactive polypeptides were visualized using horseradish peroxidase conjugated secondary antibodies (anti-rabbit or anti-mousse IgG peroxidase conjugated 1:4000; Santa Cruz Biotechnology, Dallas, TX, USA) and enhanced-chemiluminescence detection reagents (Amersham Bioscience, GE Healthcare Bio-Sciences, Pittsburgh, PA, USA) following the manufacturer-supplied protocols. Western blots were analyzed using Un-Scan-It Gel 6.0 (Gel graph Digitizing Software-Silk Scientific Corporation, UT, USA) to provide quantitative values for relative expression of each protein (normalized to its own loading control).

We found that melatonin induced a decrease in the concentration of cytosolic (Fig. 1a) and mitochondrial (Fig. 1b) Ca²⁺ in the two sarcoma cell lines tested (MG63 and sw-1353) between 4 and 24 h (Ca²⁺ levels recover after 48 h of treatment, data not shown). Intracellular ROS levels were also reduced after melatonin treatment, but this drop was only observed after 48 h of treatment (Fig. 1c).

As calcium homeostasis regulates both cell proliferation and migration, and it has been shown that melatonin regulates both processes in several types of cells, we next studied these processes in both chondrosarcoma and osteosarcoma cells. We found that changes in Ca²⁺ and ROS levels were accompanied by decreases in the numbers of viable cells after 24 and 48 h of treatment (Fig. 2a) with an increase in the number of cells in the G1 phase of the cell cycle and a decrease in the number of cells in the S phase after 24 h (Fig. 2b). The decrease in G1-S transition was confirmed by a decrease in the expression of cyclin D1 as revealed by Western blotting after 4 and 8 h of melatonin administration (Fig. 2c).

Using a scratch wound-healing assay, we found that melatonin also inhibited the migration of sarcoma cells, being more pronounced in the chondrosarcoma cells (Fig. 2d). This observation correlates well with decreases in the expression of N-cadherin and vimentin (EMT markers) after 8 h of melatonin treatment in the chondrosarcoma cells (Fig. 2e), while osteosarcoma cells showed a low basal level of N-cadherin expression (data not shown).

Mitochondrial functions such as mitochondrial respiration or ATP synthesis are also regulated by calcium homeostasis and signaling [36]. Previously, we found that melatonin decreases the expression of the E3 subunit of the pyruvate dehydrogenase complex (PDHC) in acute myeloid leukemia cells [27], which can lead to a slowdown of the Krebs cycle and, therefore, of OXPHOS. It has also been shown by others that PDHC activity / expression is regulated by Ca²⁺ concentrations [37‐39]. On the other hand, a decrease in mitochondrial Ca²⁺ may be directly related to a decrease in OXPHOS [40] and, as such, may be involved in a decrease of cell proliferation, especially if it is not accompanied by an increase in fermentative metabolism that could compensate for a decrease in oxidative energy production.

To explore whether the decrease in Ca²⁺ induced by melatonin is accompanied by changes in mitochondrial function that may be influencing cell proliferation, we first set out to study whether this indolamine modifies mitochondrial membrane potential, an indicator of mitochondrial activity [41]. We observed a decrease in mitochondrial membrane potential after 4 and 24 h of melatonin treatment (Fig. 3a). PDHC expression analysis also revealed inhibition of the expression of the E1 alpha subunit of PDHC after 4 and 8 h of treatment with the indolamine (Fig. 3b). Finally, in line with our previous results [26], no changes in fermentative metabolism, measured as LDH activity (Fig. 3c) and intracellular lactate concentration (Fig. 3d), were observed.

To corroborate the implication of the decrease in Ca²⁺ level in the inhibition of cell proliferation, cell migration, ROS levels and mitochondrial function induced by melatonin treatment, experiments were carried out supplementing Ca²⁺ or using Ca²⁺ chelators. We found that the addition of CaCl₂ to the culture medium prevented almost completely the reduction in the number of cells induced by melatonin after 48 h of treatment (Fig. 4a). Conversely, we found that addition of the Ca²⁺ chelator BAPTA reinforced the antiproliferative action of the indolamine (Fig. 4b). CaCl₂ also completely reversed the melatonin-induced inhibition of migration in both cell lines (Fig. 4c), as well as the inhibition of N-cadherin and vimentin expression induced by melatonin in chondrosarcoma cells after 8 h of treatment (Fig. 4d). Co-treatment with CaCl₂ also prevented the decrease in cellular ROS levels after 48 h of treatment (Fig. 5a). Conversely, we found that addition of the calcium chelator BAPTA to the culture medium reinforced the decrease in ROS level induced by melatonin (Fig. 5b). The decrease in mitochondrial membrane potential after 24 h of treatment with melatonin was also abolished by the addition of CaCl₂ (Fig. 5c), while it was reinforced by co-treatment with the calcium chelator BAPTA (Fig. 5d). Finally, we found that the inhibition of PDH E1 alpha expression after 8 h of treatment was also prevented by co-treatment with CaCl₂ (Fig. 5e).

The decrease in Ca²⁺ observed after melatonin administration may be due to the regulation by this indolamine of different calcium channels, either those located in the plasma membrane or in the ER. Activation of Ca²⁺ channels induces intracellular signaling cascades mediated by different pathways, such as the PI3K/Akt, MAPK, calcitonin or calcineurin pathways. Of these, the PI3K/Akt pathway has been implicated in intracellular signaling after activation of IP3R channels, while ERK is involved in signaling after activation of voltage-gated Ca²⁺ channels of the plasma membrane [42]. The expression of phosphorylated Akt and ERK as determined by Western blotting was assessed in chondro- and osteosarcoma cells after treatment with melatonin. We found that this treatment inhibited ERK phosphorylation in osteosarcoma cells after 8 and 24 h, while it had no effect on Akt activation. No effect of melatonin was observed neither on Akt nor in ERK activation in chondrosarcoma cells (Fig. 6a). The addition of CaCl₂ did not significantly abolish the inhibition of ERK activation by melatonin in osteosarcoma cells (Fig. 6b).

To study the interrelationship between the antioxidant effect of melatonin and its regulation of calcium concentration in the two human sarcoma cancer cell lines, comparative experiments were carried out using two other antioxidants. We found that the addition of antioxidants to the culture medium did not have the same effect as melatonin. Although antioxidants potentiated the antiproliferative effect of melatonin (Fig. 7a), this potentiation was independent of the regulation of Ca²⁺ levels by these molecules, since there was no effect on cytosolic calcium concentration after 24 h of treatment (Fig. 7b). Twenty-four hours of treatment with antioxidants also failed to decrease the mitochondrial membrane potential (Fig. 7c). Finally, co-treatment with trolox did not reinforce the effect of melatonin, neither on the cytosolic calcium concentration nor on the mitochondrial function (Figs.7d and e).