Curcumin-induced antiproliferative and proapoptotic effects in melanoma cells are associated with suppression of IκB kinase and nuclear factor κB activity and are independent of the B-Raf/mitogen-activated/extracellular signal-regulated protein kinase pathway and the Akt pathway
Doris R. Siwak M.S.,
Department of Experimental Therapeutics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas
Nuclear factor-κB (NF-κB) plays a central role in cell survival and proliferation in human melanoma; therefore, the authors explored the possibility of exploiting NF-κB for melanoma treatment by using curcumin, an agent with known, potent, NF-κB-inhibitory activity and little toxicity in humans.
Three melanoma cell lines (C32, G-361, and WM 266-4), all of which had B-raf mutations, were treated with curcumin, and the authors assessed its effects on viability ((3-[4,5-dimethylthiazol-2-yl]2,5-diphenyltetrazolium bromide assay) and apoptosis (flow-cytometric analysis of annexin V/propidium iodide-stained cells). Curcumin-treated cells also were examined for NF-κB binding activity (electrophoretic mobility shift assay) and for the activity of its upstream regulator, IκB kinase (IKK) (immune complex kinase assay). In addition, relevant signaling, as reflected by B-Raf kinase activity (kinase cascade assay), and steady-state levels of activated, downstream effectors, as reflected by mitogen-activated signal-regulated protein kinase (MEK), extracellular signal-regulated protein kinase (ERK), and Akt phosphorylation levels (immunoblots), were assessed.
Curcumin treatment decreased cell viability of all 3 cell lines in a dose-dependent manner (50% inhibitory concentration = 6.1–7.7 μM) and induced apoptosis. NF-κB and IKK were active constitutively in all melanoma cell lines examined, and curcumin, under apoptosis-inducing conditions, down-regulated NF-κB and IKK activities. However, curcumin did not inhibit the activities of B-Raf, MEK, or ERK, and Akt phosphorylation was enhanced. Furthermore, in the presence of curcumin, the Akt inhibitor 1L-6-hydroxymethyl-chiro-inositol 2-[(R)-2-O-methyl-3-O-octadecylcarbonate] no longer suppressed Akt phosphorylation.
Melanoma is a cutaneous tumor characterized by abnormal proliferation of melanocytes that invade the basement membrane. The majority of metastatic melanomas are resistant to diverse chemotherapeutic agents (for reviews, see Sonetas and Lowe1 and Molife and Hancock2), and long-term survival for patients with melanoma who have metastatic disease is dismal.3, 4 Consequently, the search for novel antimelanoma agents continues.
The natural compound curcumin (diferuoylmethane) has antiproliferative and proapoptotic effects against a variety of tumors in vitro (for review, see Aggarwal et al.5). These effects correlate with the inhibition of signaling proteins, including transcription factors (such as nuclear factor κB [NF-κB], activator protein-1, and early growth response-1), as well as cyclooxygenase 2, nitric oxide synthase, matrix metalloproteinase-9, urinary plasminogen activator, tumor necrosis factor, chemokines, cell surface adhesion molecules, cyclin D1, growth factor receptors (such as epidermal growth factor receptor [EGF] and human EGF receptor 2), other protein tyrosine kinases, and serine/threonine kinases.5 Despite its broad range of activity in vitro, curcumin is nontoxic in vivo; no dose-limiting toxicity was observed in individuals who ingested 8 g per day for 3 months.6
In the current study, we tested curcumin against melanoma cell lines to examine its effects on cell viability, apoptosis, and cell signaling molecules associated with melanoma pathogenicity. It has been shown previously that several signaling molecules are critical to the proliferation of melanoma in vitro or in vivo, including B-Raf (through mutation7), mitogen-activated protein kinase (MAPK),8 Akt,9 and NF-κB.10 We show that curcumin exposure at micromolar concentrations both decreased viability and induced apoptosis in a dose-dependant manner in the melanoma cell lines. Furthermore, curcumin treatment, under apoptosis-inducing conditions, attenuated NF-κB binding activity without suppressing the B-Raf kinase pathway or Akt phosphorylation in these cell lines. These results suggest that curcumin induces apoptosis in melanoma cells by inhibiting NF-κB in a B-Raf-independent and Akt-independent manner.
MATERIALS AND METHODS
The human melanoma cell lines C32, G-361, and WM 266-4 were purchased from American Type Cell Culture (Manassas, VA). All cells were grown in medium supplemented with 10% fetal bovine serum; G-361 cells were grown in McCoy 5A medium; WM 266-4 cells were grown in minimum essential medium (MEM); and C32 cells were grown in MEM supplemented with nonessential amino acids and sodium pyruvate. All media were purchased from Invitrogen Corporation (Carlsbad, CA).
The 3-[4,5-Dimethylthiazol-2-yl]2,5-Diphenyltetrazolium Bromide Cell-Viability Assay
To determine the effect of curcumin on cell viability, the 3-[4,5-dimethylthiazol-2-yl]2,5-diphenyltetrazolium bromide (MTT) assay was used. Cells were seeded 3.0–5.0 × 103 cells per well, depending on the cell line, in a 96-well plate and incubated overnight. Cells were treated in sextuplicate with 0 μM, 5 μM, 10 μM, 20 μM, or 40 μM of curcumin (Sigma-Aldrich Company, St. Louis, MO) dissolved in dimethyl sulfoxide (DMSO) and were added to their respective medium for a final DMSO concentration of 0.1%. Cells were protected from light for the duration of curcumin treatment and were incubated for 96–120 hours, depending on the cell line. Medium was aspirated from the wells; the cells were rinsed with phosphate buffered saline (PBS), and 200 μL of the MTT solution (Sigma-Aldrich Company; final concentration, 0.5 mg/mL) were added to the cells. (MTT solution was prepared by adding 1 part of stock 5-mg/mL MTT dissolved in PBS to 9 parts of serum-supplemented medium.) Cells were incubated for 4–7 hours before the medium was aspirated again, and the precipitated formazan was dissolved by adding 200 μL of DMSO and placing it on a shaker for 10–20 minutes. The samples were read at 560 nm (minus the 650-nm wavelength reference) on a plate reader (Molecular Devices Corp., Sunnyvale, CA). The mean and standard error for each treatment was determined and then was converted to the percent relative to control, with 0.1% DMSO only = 100%. The concentration at which cell growth was inhibited by 50% (the 50% inhibitory concentration [IC50]) was determined by linear interpolation [(50% − low percentage)/(high percentage − low percentage)] × (high concentration − low concentration) + low concentration]. The IC90 was determined similarly, with 90% substituting for 50% in the equation.
Cell Recovery Assay
To determine whether the effects on viability were reversible, the following cell recovery assay was performed. Cells were seeded 5.0 × 105 cells per 100-mm plate and incubated overnight. Cells were treated with 0 μM, 10 μM, 20 μM, or 40 μM curcumin dissolved in DMSO, added to medium (final DMSO concentration, 0.1%), and incubated for 96 hours. Cells were trypsinized and counted; the same number of cells for each treatment (5.0 × 103 cells per well) were seeded in sextuplicate into 96-well plates with curcumin-free medium and incubated for 96 hours. The degree of cell recovery was determined with the MTT assay.
Annexin V-Propidium Iodide Staining for Apoptotic Cells
To determine the effect of curcumin on apoptosis, cells were stained with annexin V/propidium iodide and assessed by fluorescence-activated cell sorting analysis. Cells were seeded 5.0 × 105 cells per 100-mm plate and incubated overnight. Cells were treated with 0 μM, 10 μM, 20 μM, or 40 μM curcumin in DMSO (final DMSO concentration, 0.1%) and incubated for 72 hours. Then, cells were harvested by quick (< 5 minutes) trypsinization to minimize potentially high annexin V background levels in adherent cells, and they were washed and stained with fluorescein 5(6)-isothiocyanate (FITC)-annexin V-propidium iodide, as directed on the Annexin V-FLUOS staining kit (Roche Diagnostics Corp., Indianapolis, IN). Stained cells were placed on ice and protected from light until they were read by flow cytometry. The cells were analyzed on an Epics XL-MCL flow cytometer using the System II version 3.0 software (both hardware and software from Beckman Coulter, Inc., Miami, FL) with the laser excitation wavelength at 488 nm. The green signal from FITC-annexin V was measured at 525 nm, and the red signal from propidium iodide was measured at 620 nm.
Electrophoretic Mobility Shift Assay for NF-κB Binding
Because melanoma cells often have constitutively activated NF-κB,10–12 the effect of curcumin on NF-κB binding was determined using electrophoretic mobility shift assay (EMSA). Cells seeded at 5.0–12.0 × 105 cells per 100-mm plate (depending on the cell line) and incubated overnight were treated with 0 μM or 10 μM curcumin (final DMSO concentration, 0.1%) and incubated for 72–96 hours, depending on the cell line. Cells were harvested by trypsinizing and washing twice in cold PBS, and they were stored at − 80 °C and used within 2 weeks.
Cells were thawed on ice and induced to swell by adding cold cytoplasmic extraction buffer (CEB) (10 mM N-2-hydroxyethyl piperazine-N′-2-ethane sulphonate [HEPES], pH 7.9; 10 mM KCl, 0.1 mM ethylenediamine tetraacetic acid [EDTA]; 0.1 mM ethylene guanine tetraacetic acid [EGTA], 1.0 mM dithiothreitol [DTT], 0.5 mM phenylmethanesulfonyl fluoride [PMSF]; 2 μg/mL Leupeptin; 2 μg/mL Aprotinin; and 0.5 mg/mL benzamidine) followed by a 15-minute incubation on ice. The cytoplasmic cell fraction was lysed by adding 3.13 μL/100 μL CEB of 10% Igepal (Sigma-Aldrich Company) and vortexing for 20 seconds. The cell suspension was centrifuged for 5 minutes, the cytosolic supernatant was discarded, and cold nuclear extraction buffer (20 mM HEPES, pH 7.9; 400 mM NaCl; 1.0 mM EDTA; 1.0 mM EGTA; 1.0 mM DTT; 0.5 mM PMSF; 2 μg/mL Leupeptin; 2 μg/mL Aprotinin; and 0.5 mg/mL benzamidine) was added to the pellet. Nuclear suspensions were incubated for 30 minutes on ice (vortexing every 10 minutes), then centrifuged for 10 minutes. Nuclear extracts from the supernatant were collected and stored at − 80 °C and were used within 4 days. Protein concentration was determined by the Bio-Rad DC Protein Assay (Bio-Rad Laboratories, Hercules, CA). The binding reaction was initiated by adding 20–30 μg of nuclear extract to binding buffer (100 mM HEPES, pH 7.9; 50 mM EDTA; 100 mM DTT; 10% glycerol); 2 μg poly dI:dC (Amersham Biosciences, Piscataway, NJ); 3.0 × 105 cpm of 32P-labeled, NF-κB, double-stranded oligonucleotide; and 10% Igepal (total volume, 20 μL), and incubating the mixture for 15 minutes at 37 °C. The reaction was terminated by adding 4 μL of 6 × DNA loading dye and placing samples on ice. Samples were loaded on a prerun 5.5% polyacrylamide gel and electrophoresed under nondenaturing conditions. The gel was dried and placed on film for autoradiography.
For supershifts, 2 μg of anti-p50 or p65 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were added to the nuclear extract and incubated at room temperature for 15 minutes prior to its addition in the binding reaction. Quantitation of autoradiographs was performed on the Fluorchem® 8900 Imaging System using AlphaEase® FC software (Alpha Innotech, San Leandro, CA).
The consensus sequence wild-type and mutant, blunt-end, double-stranded oligonucleotides used for the NF-κB electrophoretic mobility shifts assays were purchased from Santa Cruz Biotechnology, Inc. The NF-κB wild-type sense strand sequence was 5′-AGTTGAGGGGACTTTCCCAGGC-3′, and the mutant sequence was 5′-AGTTGAGGCGACTTTCCCAGGC-3′.13 Oligonucleotides were γ-32P-end-labeled with T4 polynucleotide kinase (New England Biolabs, Inc., Beverly, MA), purified with Quick Spin G-50 Sephadex columns (Roche Diagnostics Corp.), and stored at − 20 °C.
A kinase assay using phosphorylation of substrate was used to examine IKK activity. Cells were seeded and treated with curcumin under the conditions described above for the NF-κB EMSA. Cells were lysed in IKK lysis buffer (20 mM HEPES, pH 7.9; 50 mM NaCl; 1% Igepal; 2 mM EDTA; 0.5 mM EGTA; 2 μg/mL Aprotinin, 2 μg/mL Leupeptin; 0.5 mM PMSF; 2 mM Na3VO4), and IKK complexes were precipitated from whole cell extracts with 1 μg each of anti-IKKα and IKKα antibodies (Imgenex Corporation, San Diego, CA) overnight at 4 ° C followed by 20 μL of protein A/G-conjugated Sepharose beads (Pierce, Rockford, IL). After 2 hours, the beads were washed with IKK lysis buffer; the kinase assay was initiated upon addition of 15 μL kinase assay mixture (50 mM HEPES, pH 7.4; 20 mM MgCl2; 2 mM DTT; 20 μCi [γ-32P] adenosine triphosphate [ATP]; 10 μM unlabeled ATP) containing 2 μg of the substrate glutathione-S-transferase (GST)-IκBα (residue 1–54 of IκBα conjugated to GST). After incubation at 30 °C for 30 minutes, the reaction was terminated by the addition of 5 × sodium dodecyl sulfate (SDS) sample buffer and boiling for 5 minutes. Proteins were resolved on a 10% polyacrylamide gel under reducing conditions, the gel was dried, and the radiolabeled bands were visualized using a PhosphorImager.
To determine the total amounts of IKKα and IKKβ in each sample, immunoblotting was performed. Proteins (30 μg) from whole cell extract were resolved on a 7.5% acrylamide gel then electrotransferred to a nitrocellulose membrane. The membrane was blocked with 5% nonfat milk for 1 hour then incubated with a 1:1000 dilution of either anti-IKKα or anti-IKKβ antibody (Imgenex Corporation) for 1 hour. The membrane was washed and treated with horseradish peroxidase-conjugated, secondary antimouse immunoglobulin G antibody; then, proteins were detected by chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ).
Interleukin 8 Enzyme-Linked Immunosorbent Assay
The cytokine interleukin-8 (IL-8) has a proliferative role in melanomas,14 and its expression is regulated in part by NF-κB.15–17 Therefore, the effect of curcumin on soluble, extracellular IL-8 levels was determined using enzyme-linked immunosorbent assays (ELISAs). Cells were seeded and treated with curcumin under conditions described above for the NF-κB EMSA. After treatment, media from cells that were exposed to curcumin or to DMSO only were harvested and centrifuged to remove cellular debris. The supernatant was aliquoted and stored at − 80 °C. Samples were thawed, and the Quantikine® Human IL-8 Immunoassay kit (R&D Systems, Inc., Minneapolis, MN) was used to determine IL-8 concentration, according to the manufacturer's instructions, except that samples were read (on the Molecular Devices plate reader) at 490 nm with a wavelength correction at 560 nm. Results were calculated by generating a 4-parameter, logistic curve fit in the SOFTmax Pro version 2.6 software (Molecular Devices Corp., Sunnyvale, CA).
B-Raf Kinase Assay
Because melanoma cells often have activating B-raf mutations (see Davies et al.21), the effect of curcumin on B-Raf kinase activity was determined using the B-Raf Kinase Cascade Assay Kit (Upstate Cell Signaling Solutions, Charlottesville, VA). This kit examines Raf kinase activity by sequential phosphorylation of 0.4 μg purified mitogen-activated signal-regulated protein kinase 1 (MEK1) (the Raf substrate) and 1.0 μg extracellular signal-regulated protein kinase 2 (ERK2) (the MEK1 substrate) with nonradiolabeled ATP; 10% of this mixture is removed, and 50 μg of myelin basic protein (the ERK2 substrate) along with radioisotope-labeled ATP ([γ-32P]ATP) are added. The extent of myelin basic protein phosphorylation is determined by liquid scintillation counting. This method, due to increasing amounts of downstream components, enables the detection of Raf activity with great sensitivity. The specificity of B-Raf kinase activity is enabled by immunoprecipitation of B-Raf in cell lysates prior to the initiation of the kinase cascade assay using anti-B-Raf antibodies (Santa Cruz Biotechnology, Inc.).
Cells were seeded and treated with curcumin under the conditions described for the NF-κB EMSA. Cells were harvested and stored at − 80 °C, and they were used within 2 weeks of harvest. Upon thawing, cells were lysed with lysis buffer (137 mM NaCl; 25 mM N-Tris[hydroxymethyl]methyl-2-animoethanesulfonic acid [TES], pH 7.4; 10% glycerol; 1% Triton X-100; 0.1% Igepal; 0.01% sodium deoxycholate) supplemented with 10% protease inhibitor cocktail (Sigma-Aldrich Company), 10 mM NaF, and 0.5 mM Na3VO4, and they were passed ≈ 8 times through a 21-gauge needle. Cell lysates were centrifuged in a microcentrifuge at > 13,000 revolutions per minute for 10 minutes at 4 °C. Supernatants were collected, and the total protein concentration was determined; 1000 μg of total protein (with sample volumes equalized to 500 μL) were aliquoted for each treatment.
B-Raf was immunoprecipitated by adding 2 μg of the rabbit anti-B-Raf antibody, incubating on ice for 1.5 hours, then adding 20 μL of protein A/G-conjugated agarose beads (Oncogene Research Products, San Diego, CA), and incubating at 4 °C on a rocker for 30 minutes. For negative controls, 2 μg of rabbit antimouse antibody (ICN Pharmaceuticals, Inc., Aurora, OH) were used in place of the anti-B-Raf antibody. Immunoprecipitates were washed twice with lysis buffer supplemented with 10 mM NaF and 0.5 mM Na3VO4 and washed once with kinase assay buffer (25 mM TES, pH 7.4; 5 mM MgCl2; 5 mM MnCl2) supplemented with 10 mM NaF and 0.5 mM Na3VO4. B-Raf kinase activity was determined by the B-Raf Kinase Cascade Assay Kit (Upstate Group, Inc., Lake Placid, NY), as described above, according to manufacturer's instructions.
Immunoblots for MEK, ERK, and Akt Phosphorylation
The effects of curcumin on B-Raf/MEK/ERK signaling and on Akt activity were assessed using Western blots to examine phosphorylation of MEK, ERK, and Akt. Cells were seeded and treated with curcumin as described for the NF-κB EMSA. Cells were thawed and lysed by resuspending in B-Raf kinase assay lysis buffer plus inhibitors (10% protease inhibitor cocktail, 10 mM NaF, 0.5 mM Na3VO4) and incubating on ice for 30 minutes with vortexing every 10 minutes. Cells were then centrifuged, and the protein concentration of the supernatant was determined. Samples (40 μg of protein with sample volumes equalized to 20 μL) were aliquoted and prepared for SDS electrophoresis by adding 3 × sample buffer and boiling.
After SDS-polyacrylamide gel electrophoresis, proteins were electrotransferred from gel to nitrocellulose membrane: The membrane was blocked using 20 mM TES, pH 7.6; 137 mM NaCl; and 0.5% Tween 20 (TBS-T) plus 5% (weight/volume) nonfat dry milk (Bio-Rad Laboratories) for 30 minutes at room temperature. The membrane was rinsed, and the primary antibody diluted in TBS-T plus 1% milk was added and incubated either overnight at 4 °C or for 2–3 hours at room temperature. The primary antibodies used were antiphosphorylated (antiphospho)-MEK1/2 (dilution, 1:3000), antiphospho-ERK1/2 (dilution, 1:3000), antiphospho-Akt (dilution, 1:1000), or anti-MEK1/2 (dilution, 1:3000), anti-ERK1/2 (1:3000), or anti-Akt (dilution, 1:3000). All antibodies were from Cell Signaling Technology™ (Beverly, MA). The membrane was rinsed 3 times with TBS-T, and peroxidase-conjugated antirabbit secondary antibody diluted 1:1000–1:2000 in TBS-T and 2.5% milk was added; and the membrane was incubated at room temperature for 2.0–2.5 hours. Excess antibody was washed from the membrane, and proteins were detected by enhanced chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ).
To determine whether an Akt inhibitor would suppress Akt activity in the presence of curcumin, cells were treated with 30 μM, 45 μM, or 60 μM of Akt inhibitor (1L-6-hydroxymethyl-chiro-inositol 2-[(R)-2-O-methyl-3-O-octadecylcarbonate], Calbiochem, San Diego, CA) with and without 10 μM curcumin and were incubated for 72 hours before harvesting, followed by analysis of Akt phosphorylation levels by anti-phospho-Akt Western blots.
Prior to treating all melanoma cell lines with varying doses of curcumin, time-course experiments in which a faster growing cell line (G-361 cells, doubling time ≈ 24 hours) and a slower growing cell line (C32 cells, doubling time ≈ 33 hours) were performed to determine optimal treatment time. These cells were treated with 0–40 μM of curcumin, and effects on cell viability, as assessed by the assay, were determined every 24–48 hours for a maximum of 144 hours after treatment. Maximum dose-dependent decreases in viability were observed at 96 hours and 120 hours for G-361 cells and C32 cells, respectively, corresponding to 3–4 cell doublings for each cell line (data not shown).
To further determine the effect of curcumin on cell viability, C32, G-361, and WM 266-4 cells (doubling time ≈ 24 hours) were treated with 0–40 μM curcumin (final DMSO concentration of control and all curcumin concentrations, 0.1%), incubated for 96–120 hours (depending on the cell line), and assayed with MTT. Curcumin inhibited cell viability in all 3 melanoma cell lines in a dose-dependent manner (Fig. 1); the IC50 values in the cell lines ranged from 6.1 μM to 7.7 μM, whereas the IC90 values ranged from were 15.9 μM to 21.4 μM.
To determine whether the effects of curcumin on viability were reversible, cells were treated with 0–40 μM curcumin for 96 hours, then replated with the same number of cells in curcumin-free medium, and allowed to grow for an additional 96 hours. Cell viability was then determined by MTT recovery assay. At 10 μM curcumin, C32 cells had dramatically reduced cell recovery, whereas recovery of WM 266-4 cells was comparable to that of controls (Fig. 2). At 20 μM, cell recovery was reduced greatly in both cell lines tested. These results indicate that curcumin induces irreversible effects on viability in melanoma cells, particularly at concentrations ≥ IC50.
To determine whether curcumin induced apoptosis in melanoma cells, the cells were treated with 0–40 μM curcumin and incubated for 72 hours; this was followed by FITC-annexin V-propidium iodide staining of lightly trypsinized cells. Flow cytometric analysis revealed an increase in the proportion of annexin V-positive cells with increasing concentrations of curcumin (Fig. 3A), indicating a dose-dependent increase in apoptotic cells upon curcumin treatment. Curcumin at 10 μM increased the proportion of apoptotic cells 1.9–2.8–fold, whereas curcumin at 20 μM increased the proportion of apoptotic cells 7.2–19.7-fold compared with controls, depending on the cell line (Fig. 3B).
Curcumin Inhibits NF-κB Binding Activity in Melanoma Cells
Because the NF-κB transcription factor has been implicated in the pathogenesis of diverse malignancies, including melanoma,10 we determined whether NF-κB binding was decreased in curcumin-treated melanoma cells. Cells were treated with 0 μM or 10 μM curcumin for 72 hours and assayed for NF-κB binding with an NF-κB EMSA.
Comparison of bands from nuclear extracts that were obtained from untreated cells probed with wild-type or mutant NF-κB oligonucleotides indicated that 1 band was NF-κB-specific (Fig. 4A, lanes 1 and 2 for each cell line). Preincubation of nuclear extracts with anti-p50 or anti-p65 antibodies also reduced the intensity of this band, confirming that the band consisted of the p50/p65 heterodimers.
NF-κB was bound constitutively in the three cell lines tested. NF-κB binding activity was decreased partially in all cell lines that were treated with curcumin compared with controls (Fig. 4A,B).
Curcumin Inhibits IκB Kinase Activity in Melanoma Cells
NF-κB is activated by IκB kinase (IKK), which phosphorylates and thereby initiates signals required for degradation of the negative NF-κB regulator, IκBα (for review, see Rothwarf and Karin18). Because curcumin modulated NF-κB binding activity in melanoma cells, we examined IKK activity in curcumin-treated melanoma cells. An in vitro kinase assay using immunoprecipitated IKK from melanoma cells (either untreated or treated with 0.1% DMSO only or with 10 μM curcumin) and the GST-IκBα as substrate showed that IKK was active constitutively in the cell lines examined. This activity was decreased after curcumin treatment (Fig. 5, top). Immunoblotting of the cell extracts of untreated, 0.1% DMSO only, or curcumin-treated cells showed no significant changes in the steady-state protein levels of the IKK subunits IKKα and IKKβ in treated cells compared with cells that were untreated or exposed to vehicle only (Fig 5., middle and bottom, respectively).
IL-8 Secretion Is an Independent Factor in Curcumin-Treated Melanoma Cells
IL-8 is a cytokine in which increased serum levels are correlated with increased malignant phenotype and poor prognosis in patients with melanoma.19 IL-8 is regulated at least in part by NF-κB.20 To determine whether curcumin can modulate IL-8 secretion and whether these changes correlate with NF-κB modulation, the media of cells exposed to 0 μM or 10 μM curcumin for 72–96 hours were examined for IL-8 levels using an ELISA. After curcumin exposure, IL-8 levels increased in 2 cell lines (G-361 and WM 266-4). Baseline levels in C32 cells were low and appeared to decrease after curcumin exposure (Fig. 6). There was no clear-cut correlation between IL-8 levels and growth inhibition or apoptosis, indicating that soluble, extracellular IL-8 is not a reliable marker of curcumin-induced melanoma cell growth inhibition or apoptosis. In addition, no precise correlation was observed between curcumin-induced NF-κB inhibition and effect on IL-8, suggesting that IL-8 secretion is not dependent on NF-κB binding activity in curcumin-treated melanoma cells.
Curcumin Does Not Affect Raf, MEK, or ERK Activity in Melanoma Cells
Activating B-raf mutations are observed in the majority of melanomas.21 C32, G-361, and WM 266-4 cells have activating B-raf mutations,21 the gene product of which is activated by Ras. Activated forms of Raf reportedly activate NF-κB,22, 23 although NF-κB is activated by other pathways as well; thus, we determined the effects of curcumin treatment on B-Raf kinase activity and on the phosphorylation of MEK (the downstream substrate of B-Raf) and ERK (the downstream substrate of MEK).
Melanoma cells treated with 0 μM or 10 μM curcumin for 72–96 hours were examined for B-Raf activity by immunoprecipitating B-Raf in these cells and using a Raf kinase cascade assay kit. This kit examines Raf kinase enzymatic activity by the sequential phosphorylation of increasing amounts of MEK1, ERK2, and myelin basic protein. Myelin basic protein phosphorylation is detected by radioisotope labeling. All three cell lines had significant baseline levels of B-Raf kinase activity (probably due to activating B-raf mutations21). Changes in B-Raf kinase activity upon exposure to curcumin were insignificant (from + 3% to − 10%) (Fig. 7A).
Melanoma cells treated with 0 μM or 10 μM curcumin for 72 hours also were assayed for the extent of MEK phosphorylation using anti-phospho-MEK1/2 immunoblots. Endogenous MEK phosphorylation was detected readily in the melanoma cell lines (Fig. 7B, top), indicating that MEK is active constitutively in melanoma cells. However, no differences were observed between control and curcumin-treated melanoma cells in the cell lines examined.
ERK1/2 reportedly activates NF-κB in melanoma cells.24 Furthermore, ERK1/2 is a downstream effector in the Raf/MEK pathway.25–27 Therefore, we also determined whether ERK1/2 activity is affected by curcumin treatment using antiphospho-ERK1/2 immunoblot. With regard to MEK, endogenous levels of ERK1/2 phosphorylation were detected readily in the melanoma cell lines (Fig. 7B, middle), indicating that ERK1/2 also is active constitutively in melanoma cells. Curcumin exposure did not affect ERK1/2 phosphorylation. Taken together, these results indicate that, although the B-Raf/MEK/ERK pathway is active constitutively in B-raf-mutated melanoma cells, curcumin does not inhibit this pathway and that NF-κB suppression is independent of this pathway.
Akt Activity Increased in Melanoma Cells with Exposure to Curcumin
The activation of Akt-mediated pathways induces NF-κB activation in different cell types.28–31 In addition, several reviews have established that the phosphatidylinositol 3-kinase (PI3K)/Akt pathway engages in crosstalk with the Ras/Raf/MEK/ERK pathway.32–34 To determine whether curcumin modulates Akt activity, melanoma cells treated with 0 μM or 10 μM of curcumin for 72–96 hours were assessed for phospho-Akt levels using an antiphospho-Akt immunoblot. Low levels of phosphorylated Akt were detected in the cell lines (Fig. 7B, bottom). However, 3.9-fold and 1.7-fold increases in phospho-Akt levels, as determined by densitometry, were observed in G-361 and WM 266-4 cells, respectively, upon curcumin treatment. Total Akt levels remained unchanged in these cells.
Based on these results, it appears that the antiproliferative/proapoptotic effect of curcumin was independent of Akt. However, it remains possible that retention and enhancement of Akt operates as a salvage pathway. Therefore, we used the Akt inhibitor 1L-6-hydroxymethyl-chiro-inositol 2-[(R)-2-O-methyl-3-O-octadecylcarbonate] and first determined its effect alone on cell proliferation/survival and Akt activity on G-361 cells. The IC50 for cell proliferation/survival of the Akt inhibitor, as determined by the MTT assay, 61 μM (data not shown)l whereas Akt phosphorylation, as determined by Western blot analysis, was inhibited with an IC50 of ≈ 58 μM. However, when 10 μM of curcumin were added in combination with 60 μM of the Akt inhibitor, no decrease in Akt phosphorylation was observed (Fig. 8).
In the current study, we demonstrated that curcumin inhibited cell viability and induced apoptosis in a dose-dependent manner in melanoma cells (Figs. 1, 3). Furthermore, we showed that NF-κB activity, as measured by EMSA, was inhibited in part under conditions in which melanoma cells were induced to undergo apoptosis (72 hours of exposure to 10 μM of curcumin) (Fig. 4). Previous reports described NF-κB inhibition with exposure to higher concentrations of curcumin (60 μM) for shorter periods (6 hours) in melanoma cell lines.35 Therefore, the NF-κB machinery is suppressed both by short exposures to high concentrations of curcumin and by longer exposures to lower concentrations of curcumin. In addition, we demonstrated that, under apoptosis-inducing conditions, IKK, the upstream regulator of NF-κB, is inhibited strongly by curcumin (Fig. 5). Partial inhibition of NF-κB but strong inhibition of IKK by curcumin suggests that, in melanoma cells, signaling molecules other than IKK can regulate NF-κB activity. Indeed, IKK-independent NF-κB activation has been demonstrated previously in ultraviolet-irradiated cells,36, 37 in nontypeable Hemophilus influenzae-infected cells,38 and in mutant cultured human cells with constitutively activated NF-κB.39 In melanoma cells, there is evidence that an ERK1/2-mediated pathway and an Akt-mediated pathway can contribute to NF-κB activation independent of IKK.24, 40
To determine whether other potential NF-κB regulators also are involved in curcumin-induced inhibition of proliferation or survival in melanoma cells, we examined the B-Raf/ERK kinase pathway as well as Akt activity. Both the Ras/ERK pathway, of which B-Raf is a signaling intermediate, and the PI3K/Akt pathway play a role in melanoma cell proliferation or survival41–43 in addition to regulating NF-κB activity.22, 28, 44–45 B-Raf is especially important in this regard, because activating B-raf mutations are a hallmark of melanomas.21 In the current investigation, we showed that, under apoptosis-inducing conditions, neither the B-Raf/ERK pathway nor the Akt pathway was inhibited upon exposure to curcumin. In fact, Akt phosphorylation was increased after curcumin treatment (Fig. 7B). These results are in contrast to those seen with curcumin-treated breast, prostate, and renal carcinoma cells, all of which showed decreased Akt activity upon exposure to curcumin.46–48 Therefore, it is possible that the induction of apoptosis in melanoma cells either does not require suppression of these pathways, or persistent activation of these signals eventually emerges as an escape pathway that could lead to chemoresistance in melanoma cells. The latter may be of concern. because we demonstrated that the Akt inhibitor 1L-6-hydroxymethyl-chiro-inositol 2-[(R)-2-O-methyl-3-O-octadecylcarbonate] was unable to suppress Akt phosphorylation in the presence of curcumin (Fig. 8). Still, the implications of these experiments for the antitumor potential of curcumin is unclear, because the role of Akt in preventing apoptosis in melanoma cells remains controversial, with some experiments reporting that Akt inhibition induces apoptosis49 and with others suggesting that Akt inhibition induces cell cycle arrest, but not programmed cell death.9 In addition, curcumin retains its antiproliferative effects in our melanoma cells despite the enhancement of Akt phosphorylation (Fig. 8). With regard to ERK, or MAPK, Smalley and Eisen reported that the inhibition of MAPK did not result in apoptosis of melanoma cells,9 and the current results suggesting that apoptosis is independent of MAPK suppression are in agreement with their report. Finally, based on the ability of ERK1/2 or Akt to activate NF-κB independent of IKK,24, 40 the lack of inhibition of ERK1/2 and Akt in our melanoma cells after curcumin treatment may explain the differences in inhibition observed between IKK activity and NF-κB activity.
IL-8 stimulates melanoma cell proliferation in vitro,14 and increased levels of this cytokine correlate with a poor prognosis in patients with melanoma.19 The promoter region of IL-8 contains a recognition site for the NF-κB transcription factor.20 In this report, we also explored the consequence of NF-κB inactivation by assessing IL-8 secretion levels. It was somewhat surprising that IL-8 levels increased after curcumin exposure in the high IL-8-secreting melanoma cell lines (Fig. 6). Therefore, it appears that IL-8 secretion is independent of curcumin-induced NF-κB inhibition. This may be because IL-8 expression also is regulated transcriptionally by AP-1 and C/EBP as well as NF-κB (reviewed by Hoffmann et al.50). AP-1 activity is regulated by the ERK and Akt pathways, as reported in several reviews.51–54 Whereas the ERK pathway is preserved after curcumin exposure, it is possible that AP-1, through activation of the Akt pathway, induces IL-8 transcription in melanoma cells in the presence of curcumin-induced NF-κB suppression.
Based on our studies, we conclude that curcumin is a potent suppressor of cell viability and inducer of apoptosis in melanoma cell lines. This activity is associated with inhibition of the IKK/NF-κB transcriptional machinery, but not the B-Raf/ERK or Akt pathways. Indeed, curcumin enhances Akt activity and results in resistance to Akt suppression by the Akt inhibitor 1L-6-hydroxymethyl-chiro-inositol 2-[(R)-2-O-methyl-3-O-octadecylcarbonate]. These results imply that suppression of the viability of melanoma cells can occur despite the continued activation of the B-Raf/MEK/ERK and Akt pathways. Future investigations to determine the effects of curcumin in animal models of melanoma and clinical trials are planned.
The authors thank Karen Ramirez for her flow cytometric work in the annexin V-propidium iodide cell staining experiments.